JP2004015819A - Method and apparatus for estimating channel, method and apparatus for demodulation, and method and apparutus for determining fading frequency - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method and an apparatus for determining a fading frequency on the basis of the inner product value of a pilot symbol. <P>SOLUTION: The method calculates the inner product value of the pilot symbol multiplexed in terms of timing by a multiplexed parallel control channel for a data channel, and determines the fading frequency on the basis of the inner product value. That means, for the calculation of the inner product values, the method calculates the inner product value for the average value of two normalized pilot symbols by normalizing the average value of the pilot symbols of two slots for each slot in the control channel. The method averages the calculated inner products over a plurality of slots formed in the control channel. Furthermore, the method determines the fading frequency by comparing the averaged inner product value and a threshold. <P>COPYRIGHT: (C)2004,JPO

Description

The present invention relates to a channel estimation device and method, a demodulation device and method, and a fading frequency determination device and method. More specifically, the present invention relates to a channel estimation device, a demodulation device, and the like that can be applied to a mobile communication system that performs voice / data transmission in a high-speed fading environment. The present invention also relates to a demodulator and a demodulation method conforming to the CDMA system for performing multiple access by spreading a wideband signal with a high-speed spreading code equal to or higher than an information rate.

振幅 In a mobile communication environment, amplitude and phase fluctuations occur due to Rayleigh fading due to relative position movement between a mobile station and a base station. In a phase modulation method of transmitting information at a carrier phase, a method of identifying and determining information data by differentially encoding and placing information on the relative phase of the preceding and succeeding symbols and performing delay detection on the receiving side is generally used. It was a target. However, in this differential detection, since the transmission data is differentially encoded as described above, a one-bit error in a wireless section becomes a two-bit error in information data. For this reason, in the two-phase phase modulation method (BPSK modulation), for example, the reception error rate is deteriorated by 3 dB for the same signal power to interference / noise power ratio (SNIR) as compared with the synchronous detection.

絶 対 Also, although the absolute synchronous detection for discriminating the phase of the received signal by the absolute phase for each symbol has a highly efficient reception characteristic, it is difficult to determine the received absolute phase in a Rayleigh fading environment.

In order to solve this problem, a method has been proposed in which a pilot symbol is inserted between data symbols and channel estimation of the data symbol is performed using the pilot symbol. As a pilot symbol insertion method, for example, there is a method of time-multiplexing a data symbol and a pilot symbol into one channel (time multiplexing method) (FIG. 16). The following documents 1 to 3 propose a channel estimation method using this time multiplexing method. Document 1 (Transactions of the Institute of Electronics, Information and Communication Engineers, Vol. J72-B-11, No. 1, pp. 7-15, pp. 7-15) In January 1989, Sanbe "Fading distortion compensation of 16QAM for land mobile communication"), in order to solve the above-mentioned problem, a fading distortion is used by using a known phase pilot symbol inserted between data symbols (information symbols) at a fixed period. Have been proposed for estimating and compensating. In this method, one pilot symbol having a known transmission phase is inserted into the communication channel for every several symbols of the data symbol, and the transmission path is estimated based on the reception phase of the pilot symbol. By measuring the amplitude and phase of the received signal of each path of each communicator in pilot symbols before and after the corresponding data symbol section, and by interpolating the measured values, the transmission path fluctuation in the data symbol section is estimated, Compensate.

In Reference 2 (IEICE Technical Report RCS97-74, Ando et al., "RAKE reception using multi-slot weighted averaging channel estimation method for pilot symbols in DS-CDMA), channel estimation is performed using more pilot symbols. A method of performing more accurate channel estimation by performing the channel estimation is performed using pilot symbols inserted at a fixed period between data symbols. In a plurality of slots before and after the slot to which the data symbol belongs, the pilot symbols (estimated complex fading envelope) are averaged (in-phase addition), and the average value is weighted and averaged by a weighting coefficient. By obtaining the channel estimate . This improves the channel estimation accuracy with respect to thermal noise and the self station multipath interference and channel interference.

In Reference 3 (IEICE Technical Report RCS98-20, Abeta et al., “Characteristics of DS-CDMA Adaptive Plural Symbol Weighted Averaging Pilot Channel Estimation Scheme”), thermal noise is reduced by adaptively controlling the weighting factor. A method has been proposed that achieves both the reduction effect and the followability to fading fluctuation. In this method, weighted averaging is used for channel estimation, and the weighting factor is sequentially changed by an adaptive signal to obtain an optimal weighting factor.

As a pilot symbol insertion method, there are a parallel time multiplexing method (FIG. 1) and a parallel method (FIG. 22) in which pilot symbols are time-multiplexed on a control channel parallel-multiplexed with a data channel, in addition to the time multiplexing method.

Also in the parallel time multiplexing method, it is desired to perform highly accurate channel estimation by calculating the channel estimation value of the data symbol of the data channel by weighting and averaging the pilot symbols.

In addition, in the methods of the above-mentioned references 1 to 3, the channel fluctuation in each slot is assumed to be small, and the same channel estimation value is obtained using the same pilot symbol for all data symbols in one slot. Therefore, there is a problem that the characteristics are deteriorated during high-speed fading.

Furthermore, in the method of Reference 2, the weighting factor is fixedly given. If the weighting factor of a slot located at a position temporally away from the slot is increased in order to reduce the effect of thermal noise, the tracking performance with respect to fading fluctuation is increased. Is deteriorated, and as a result, there is a problem that channel estimation accuracy is deteriorated. Further, although the method of Reference 3 solves the problem of Reference 2, there is a problem that the configuration of the demodulator becomes complicated by using adaptive signal processing.

By the way, in a mobile communication environment, amplitude fluctuations and phase fluctuations occur due to Rayleigh fading accompanying the movement of the relative position between the mobile station and the base station. As a method of compensating for the amplitude fluctuation and the phase fluctuation and effectively combining multipaths, a synchronous detection processing using a pilot signal is known.

で は In this method, the transmitting side transmits a known pilot signal, and the receiving side demodulates the pilot signal and averages it over time to perform channel estimation. Then, by performing phase correction of the data signal using the estimated channel vector and performing RAKE combining, demodulation using the power of the received signal effectively can be realized.

Since the channel estimation accuracy directly affects the data quality, it is necessary to perform averaging over an appropriate time section using an appropriate weight sequence. Heretofore, one series has been used as a weight series that improves channel estimation accuracy.

When channel estimation is performed on the receiving side, averaging the pilot signal using an appropriate weight sequence improves channel estimation accuracy and enables high-quality communication. The weight sequence differs depending on the propagation conditions, mainly the moving speed.

That is, when the moving speed is slow, the channel fluctuation becomes slow, so that a weight sequence that makes the averaging time longer can be effective. On the other hand, when the moving speed is fast, the weight sequence follows the fast channel fluctuation. Therefore, a weight sequence in which the averaging time is reduced to some extent is effective.

However, in the conventionally known channel estimation using only one series of weight sequences, averaging suitable for all moving speeds cannot be performed, so that communication quality deteriorates, transmission power increases, and communication capacity increases. Was causing the decline.

と し て Further, as a method of changing the weight sequence according to the moving speed, there is a method of detecting the moving speed and changing the weight sequence according to the detected speed. However, this method has a problem that if the speed detection accuracy or the detection tracking ability is poor, it is not possible to improve the communication quality, reduce the transmission power, and increase the capacity.

An object of the present invention is to perform high-precision channel estimation by calculating the channel estimation value of a data symbol of a data channel by weighting and averaging pilot symbols in a parallel time multiplexing method.

Further, a data symbol in a slot is divided into a plurality of data symbol sections, a pilot symbol appropriate for calculating a channel estimation value of the data symbol in each data symbol section is selected, and the pilot symbol is weighted and averaged to obtain a data symbol. High-precision channel estimation is performed by calculating a channel estimation value of a data symbol in a section.

Furthermore, the fading frequency is determined based on the inner product value of the pilot symbols. Another object of the present invention is to realize optimal channel estimation for fading frequencies with a simpler configuration.

Further, by directly determining and using a weight sequence effective for various moving speeds from reception quality, it is possible to improve communication quality, reduce transmission power, and increase communication capacity.

In order to achieve the above object, an invention according to claim 1 is a channel estimation device, which is a weighting device for weighting and averaging pilot symbols time-multiplexed on a control channel parallel-multiplexed with a data channel. Weighting coefficient generating means for generating coefficients, and channel estimation value calculating means for calculating a channel estimation value of a data symbol of the data channel by weighting and averaging the pilot symbols using the weighting coefficients. I do.

The invention according to claim 2 is the channel estimating device according to claim 1, wherein the weighting coefficient generation means performs weighted averaging of an average value of pilot symbols in each of a plurality of slots of the control channel. Wherein the channel estimation value calculation means calculates the channel estimation value of the data symbol of the data channel by weighting and averaging the average value of the pilot symbols using the weighting factor. .

The invention according to claim 3 is the channel estimation device according to claim 1 or 2, wherein the weighting coefficient is determined according to a position of the pilot symbol in a slot of the control channel. And

According to a fourth aspect of the present invention, in the channel estimation device according to any one of the first to third aspects, the weighting coefficient generation unit converts the data symbols in the slot of the data channel into a plurality of data symbol sections. Dividing, selecting a pilot symbol suitable for calculating a channel estimation value of a data symbol in each data symbol section, generating a weighting coefficient for averaging the pilot symbols with weight, and the channel estimation value calculating means includes: The pilot symbol is weighted and averaged using a weighting coefficient, and a channel estimation value of a data symbol in each data symbol section is calculated.

The invention according to claim 5 is the channel estimating apparatus according to claim 4, wherein the weighting coefficient generation means is configured to perform channel estimation of a data symbol in a last data symbol section of an i-th (i: integer) slot. For calculating the value and calculating the channel estimation value of the data symbol in the first data symbol section of the (i + 1) -th slot, select the same pilot symbol and generate a weighting coefficient for weighting and averaging the pilot symbol. It is characterized by doing.

The invention according to claim 6 is the channel estimation apparatus according to any one of claims 1 to 5, wherein the fading frequency determination unit determines a fading frequency based on an inner product value of the pilot symbol, and the fading frequency determination. And a coefficient changing means for changing a coefficient used for the weighted averaging in accordance with the fading frequency determined by the means.

The invention according to claim 7 is the channel estimation device according to any one of claims 1 to 6, wherein a transmission rate of the data channel is different from a transmission rate of the control channel.

The invention according to claim 8 is a demodulation device, wherein a weighting coefficient generating means for generating a weighting coefficient for weighting and averaging pilot symbols time-multiplexed on a control channel parallel-multiplexed with a data channel, A weighted average of the pilot symbols using the weighting coefficients, channel estimation value calculation means for calculating a channel estimation value of the data symbol of the data channel, and a channel estimation value calculated by the channel estimation value calculation means. Channel fluctuation compensating means for compensating for channel fluctuations of data symbols.

The invention according to claim 9 is a fading frequency determination apparatus, wherein an inner product value calculating means for calculating an inner product value of a pilot symbol time-multiplexed on a control channel parallel-multiplexed with a data channel, and the inner product value calculation. Determining means for determining the fading frequency based on the inner product value calculated by the means.

The invention according to claim 10 is the fading frequency determination apparatus according to claim 9, wherein the inner product value calculating means normalizes an average value of pilot symbols in each of two slots of the control channel. Means for calculating an inner product value of an average value of two pilot symbols normalized by the normalizing means, and an inner product value calculated by the inner product value calculating means for calculating the inner product value of the control channel. Inner product value averaging means for averaging over a plurality of slots, wherein the determining means compares the inner product value averaged by the inner product value averaging means with a threshold to determine a fading frequency. It is characterized by having.

The invention according to claim 11 is the fading frequency determination device according to claim 10, wherein when the inner product value averaged by the inner product value averaging means is larger than a certain value, the control is performed. The normalization, the dot product calculation and the dot product averaging are performed on the average value of the pilot symbols in each of the two more spaced slots of the channel, and the resulting averaged dot product and the farther interval are obtained. The fading frequency is determined by comparing the threshold value with the threshold value.

The invention according to claim 12 is the fading frequency determination apparatus according to claim 9, wherein the inner product value calculating means is configured to calculate each of two slots of the control channel for each of multipaths used for RAKE combining. Normalizing means for normalizing the average value of the pilot symbols in, and, for each of the multipaths, inner product value calculating means for calculating the inner product value of the average value of the two pilot symbols normalized by the normalizing means. A first inner product value averaging means for averaging the inner product value of each of the multipaths calculated by the inner product value calculation executing means; and controlling the inner product value averaged by the first inner product value averaging means. Second inner product value averaging means for averaging over a plurality of slots of the channel, wherein the determining means performs averaging by the second inner product value averaging means. And having a determination executing means the fading frequency was by comparing the inner product value with the threshold.

The invention according to claim 13 is the fading frequency determination device according to claim 12, wherein, when the inner product value averaged by the second inner product value averaging means is larger than a certain value, For the average value of the pilot symbols in each of the two more spaced slots of the control channel, normalize, calculate the inner product value, average the inner product value of each of the multipaths, and calculate the inner product value over the plurality of slots. Averaging is performed, and a fading frequency is determined by comparing the obtained averaged inner product value with a threshold value corresponding to the farther interval.

The invention according to claim 14 is the fading frequency determination apparatus according to claim 9, wherein the inner product value calculation means normalizes an average value of pilot symbols in each of two slots of the control channel. Means for calculating the inner product value of the average value of the two pilot symbols normalized by the normalizing means, and calculating the inner product value at least two times by changing the inner product measurement interval. Inner product value averaging means for averaging the inner product value calculated by the inner product value calculation executing means over a plurality of slots of the control channel, wherein the determination means comprises: an inner product value averaged by the inner product value averaging means. It is characterized by having a judgment execution means for judging the fading frequency using the inner product value of the measurement interval.

According to a fifteenth aspect of the present invention, there is provided the fading frequency determination device according to the fourteenth aspect, wherein a difference between the inner product measurement intervals of the two inner product measurement intervals averaged by the inner product value averaging means is calculated. Means for judging the fading frequency using the difference calculated by the difference calculation means.

The invention according to claim 16 is the fading frequency determination apparatus according to claim 9, wherein the inner product value calculating means is configured to calculate each of the two slots of the control channel for each of the multipaths used for RAKE combining. And a normalization means for normalizing the average value of the pilot symbols in each of the multipaths, and for each of the multipaths, the inner product value of the average value of the two pilot symbols normalized by the normalization means, Inner product value calculation executing means for calculating at least one inner product value; first inner product value averaging means for averaging the inner product value of each of the multipaths calculated by the inner product value calculating execution means for each inner product measurement interval; A second step of averaging the inner product value averaged by the first inner product value averaging means over a plurality of slots of the control channel for the measurement interval. Product value averaging means, wherein the determination means has determination execution means for determining a fading frequency using an inner product value for each inner product measurement interval averaged by the second inner product value averaging means. It is characterized by.

The invention according to claim 17 is the fading frequency determination device according to claim 16, wherein a difference between inner product values for two inner product measurement intervals averaged by the second inner product value averaging means is calculated. The image processing apparatus further includes a difference calculation unit, wherein the determination execution unit determines the fading frequency also using the difference calculated by the difference calculation unit.

An invention according to claim 18 is a channel estimating apparatus that calculates a channel estimation value of the data symbol by using a pilot symbol in a channel in which the data symbol and the pilot symbol are time-multiplexed. Weighting for dividing a data symbol into a plurality of data symbol sections, selecting a pilot symbol appropriate for calculating a channel estimation value of the data symbol in each data symbol section, and generating a weighting coefficient for weighting and averaging the pilot symbols A coefficient generation means, and a channel estimation value calculation means for weighting and averaging the pilot symbols using the weighting coefficients and calculating a channel estimation value of a data symbol in each data symbol section.

The invention according to claim 19 is the channel estimating device according to claim 18, wherein the weighting coefficient generation means is configured to perform channel estimation of a data symbol in a last data symbol section of an i-th (i: integer) slot. For calculating the value and calculating the channel estimation value of the data symbol in the first data symbol section of the (i + 1) -th slot, select the same pilot symbol and generate a weighting coefficient for weighting and averaging the pilot symbol. It is characterized by doing.

The invention according to claim 20 is the channel estimation apparatus according to claim 18 or 19, wherein the weighting coefficient generating means weights and averages an average value of pilot symbols in each of a plurality of slots of the channel. Generating a weighting coefficient for calculating a channel estimation value of the data symbol in each data symbol section by weighting and averaging the average value of the pilot symbols using the weighting coefficient. And

The invention according to claim 21 is the channel estimation apparatus according to any one of claims 18 to 20, wherein the weighting coefficient is determined according to a position of the pilot symbol in a slot of the channel. It is characterized.

The invention according to claim 22 is the channel estimating apparatus according to any one of claims 18 to 21, wherein the fading frequency determining means for determining a fading frequency based on an inner product value of the pilot symbols, and the fading frequency determination And a coefficient changing means for changing a coefficient used for the weighted averaging in accordance with the fading frequency determined by the means.

The invention according to claim 23 is a demodulation device, wherein a data symbol in a slot of a channel in which a data symbol and a pilot symbol are time-multiplexed is divided into a plurality of data symbol sections, and a data symbol of each data symbol section is divided. Selecting a pilot symbol appropriate for the calculation of the channel estimation value, weighting coefficient generating means for generating a weighting coefficient for weighting and averaging the pilot symbol, and weighting and averaging the pilot symbol using the weighting coefficient, Channel estimation value calculation means for calculating a channel estimation value of a data symbol in each data symbol section; and channel fluctuation compensation means for compensating for channel fluctuation of the data symbol using the channel estimation value calculated by the channel estimation value calculation means. Characterized by having

The invention according to claim 24 is a fading frequency determination apparatus, wherein: an inner product value calculating unit that calculates an inner product value of a pilot symbol in a channel in which a data symbol and a pilot symbol are time-multiplexed; Determining means for determining a fading frequency based on the calculated inner product value.

The invention according to claim 25 is the fading frequency determination device according to claim 24, wherein the inner product value calculation means normalizes an average value of pilot symbols in each of two slots of the channel. Means, an inner product value calculation executing means for calculating an inner product value of an average value of two pilot symbols normalized by the normalizing means, and an inner product value calculated by the inner product value calculating means is stored in a plurality of slots of the channel. Inner product value averaging means for averaging over, and the determining means has a determination executing means for determining a fading frequency by comparing the inner product value averaged by the inner product value averaging means with a threshold value. It is characterized.

The invention according to claim 26 is the fading frequency determination device according to claim 25, wherein when the inner product value averaged by the inner product value averaging means is larger than a certain value, the control is performed. The normalization, the dot product calculation and the dot product averaging are performed on the average value of the pilot symbols in each of the two more spaced slots of the channel, and the resulting averaged dot product and the farther interval are obtained. A fading frequency judging device which judges a fading frequency by comparing a threshold value according to the threshold value.

The invention according to claim 27 is the fading frequency determination apparatus according to claim 24, wherein the inner product value calculating means is configured to calculate each of two slots of the control channel for each of multipaths used for RAKE combining. Normalizing means for normalizing the average value of the pilot symbols in, and, for each of the multipaths, inner product value calculating means for calculating the inner product value of the average value of the two pilot symbols normalized by the normalizing means. A first inner product value averaging means for averaging the inner product value of each of the multipaths calculated by the inner product value calculation executing means; and an inner product value averaged by the first inner product value averaging means to the channel. And a second inner product value averaging means for averaging over a plurality of slots, wherein the determining means performs averaging by the second inner product value averaging means. Comparing the inner product value with the threshold value and having a determination executing means fading frequency.

The invention according to claim 28 is the fading frequency determination device according to claim 27, wherein, when the inner product value averaged by the second inner product value averaging means is larger than a certain value, For the average value of the pilot symbols in each of the two more spaced slots of the control channel, normalize, calculate the inner product value, average the inner product value of each of the multipaths, and calculate the inner product value over the plurality of slots. Averaging is performed, and a fading frequency is determined by comparing the obtained averaged inner product value with a threshold value corresponding to the farther interval.

The invention according to claim 29 is the fading frequency determination apparatus according to claim 24, wherein the inner product value calculation means normalizes an average value of pilot symbols in each of two slots of the channel. Means, inner product value calculation executing means for calculating two or more inner product values of the average values of the two pilot symbols normalized by the normalizing means at different inner product measurement intervals, and for each inner product measurement interval, the inner product Inner product value averaging means for averaging the inner product value calculated by the value calculation executing means over a plurality of slots of the control channel, wherein the determining means measures each inner product value averaged by the inner product value averaging means. It is characterized by having a judgment execution means for judging the fading frequency using the inner product value of the interval.

The invention according to claim 30 is the fading frequency determining apparatus according to claim 29, wherein the difference calculation for calculating the difference between the inner product values for the two inner product measurement intervals averaged by the inner product value averaging means. Means for judging the fading frequency using the difference calculated by the difference calculation means.

The invention according to claim 31 is the fading frequency determination device according to claim 24, wherein the inner product value calculation means is configured to determine, for each of the multipaths used for RAKE combining, each of the two slots of the channel. Normalizing means for normalizing the average value of pilot symbols; and, for each of the multipaths, two inner product values of the average values of the two pilot symbols normalized by the normalizing means. Inner product value calculation executing means for calculating the above, first inner product value averaging means for averaging the inner product value of each of the multipaths calculated by the inner product value calculating means for each inner product measurement interval, and each inner product measurement For the interval, a second inner product for averaging the inner product value averaged by the first inner product value averaging means over a plurality of slots of the control channel. Value averaging means, and the determination means has a determination execution means for determining a fading frequency using an inner product value for each inner product measurement interval averaged by the second inner product value averaging means. Features.

The invention according to claim 32 is the fading frequency determination device according to claim 31, wherein the difference between the inner product values of the two inner product measurement intervals averaged by the second inner product value averaging means is calculated. The image processing apparatus further includes a difference calculation unit, wherein the determination execution unit determines the fading frequency also using the difference calculated by the difference calculation unit.

The invention according to claim 33, comprising a channel estimation apparatus for calculating a channel estimation value of a data symbol of the data channel using pilot symbols of a pilot channel multiplexed in parallel with the data channel, wherein the data symbol of the data channel is Is divided into a plurality of data symbol sections, a pilot symbol suitable for calculating a channel estimation value of a data symbol in each data symbol section is selected, and a weighting coefficient generation for generating a weighting coefficient for weighting and averaging the pilot symbols is performed. Means, and a channel estimation value calculation means for weighting and averaging the pilot symbols using the weighting coefficients and calculating a channel estimation value of a data symbol in each data symbol section.

The invention according to claim 34 is the channel estimating apparatus according to claim 33, wherein the weighting coefficient generating means performs weighted averaging of an average value of pilot symbols in each of a plurality of sections of the pilot channel. Wherein the channel estimation value calculation means weights and averages the average value of the pilot symbols using the weighting factors, and calculates a channel estimation value of a data symbol in each data symbol section. Channel estimation device.

A thirty-fifth aspect of the present invention is the channel estimation apparatus according to the thirty-third or thirty-fourth aspect, wherein the fading frequency determination means determines a fading frequency based on an inner product value of the pilot symbol, and the fading frequency determination means And a coefficient changing means for changing a coefficient used for the weighted averaging in accordance with the fading frequency determined by the means.

The invention according to claim 36 is the channel estimation apparatus according to any one of claims 33 to 35, wherein a transmission rate of the data channel is different from a transmission rate of the pilot channel.

38. The demodulation apparatus according to claim 37, wherein the data channel of the data channel is divided into a plurality of data symbol sections, and the data channel is suitable for calculating a channel estimation value of the data symbol in each data symbol section. Selecting pilot symbols of pilot channels multiplexed in parallel to each other, generating weighting coefficients for weighting and averaging the pilot symbols, weighting coefficient generation means, and weighting and averaging the pilot symbols using the weighting coefficients, Channel estimation value calculation means for calculating a channel estimation value of a data symbol in a data symbol section; and channel fluctuation compensation means for compensating for channel fluctuation of the data symbol using the channel estimation value calculated by the channel estimation value calculation means. It is characterized by having.

The invention according to claim 38, wherein the fading frequency determination device, the inner product value calculating means for calculating the inner product value of the pilot symbols of the pilot channel parallel multiplexed to the data channel, the inner product calculated by the inner product value calculating means Determining means for determining the fading frequency based on the value.

The invention according to claim 39 is the fading frequency determination apparatus according to claim 38, wherein the inner product value calculation means normalizes an average value of pilot symbols in each of two sections of the pilot channel. Means for calculating an inner product value of an average value of two pilot symbols normalized by the normalizing means; and an inner product value calculated by the inner product value calculating means, Inner product value averaging means for averaging over a section, and the determination means has a determination execution means for determining a fading frequency by comparing the inner product value averaged by the inner product value averaging means with a threshold value It is characterized by the following.

The invention according to claim 40 is the fading frequency determination device according to claim 39, wherein, when the inner product value averaged by the inner product value averaging means is larger than a certain value, the pilot The normalization, the inner product value calculation and the inner product value averaging are performed on the average value of the pilot symbols in each of the two farther intervals of the channel, and the obtained averaged inner product value and the farther interval are obtained. The fading frequency is determined by comparing the threshold value with the threshold value.

The invention according to claim 41 is the fading frequency determination apparatus according to claim 38, wherein the inner product value calculating means is configured to calculate each of the two sections of the pilot channel for each of the multipaths used for RAKE combining. Normalizing means for normalizing the average value of the pilot symbols in, and, for each of the multipaths, inner product value calculating means for calculating the inner product value of the average value of the two pilot symbols normalized by the normalizing means. A first inner product value averaging means for averaging the inner product value of each of the multipaths calculated by the inner product value calculation executing means, and an inner product value averaged by the first inner product value averaging means as the pilot A second inner product value averaging means for averaging over a plurality of sections of the channel, wherein the determining means comprises: And having a determination executing means the fading frequency by comparing the disproportionation inner product value with the threshold value.

The invention according to claim 42 is the fading frequency determination device according to claim 41, wherein when the inner product value averaged by the second inner product value averaging means is larger than a certain value, For the average value of the pilot symbols in each of the two more spaced intervals of the pilot channel, normalize, calculate the inner product value, average the inner product value of each of the multipaths, and calculate the inner product value over the plurality of intervals. Averaging is performed, and a fading frequency is determined by comparing the obtained averaged inner product value with a threshold value corresponding to the farther interval.

The invention according to claim 43 is the fading frequency determination apparatus according to claim 38, wherein the inner product value calculating means normalizes an average value of pilot symbols in each of two sections of the pilot channel. Means for calculating the inner product value of the average value of the two pilot symbols normalized by the normalizing means, and calculating the inner product value at least two times by changing the inner product measurement interval. Inner product value averaging means for averaging the inner product value calculated by the inner product value calculation executing means over a plurality of sections of the control channel, wherein the determination means comprises: an inner product value averaged by the inner product value averaging means. It is characterized by having a judgment execution means for judging the fading frequency using the inner product value of the measurement interval.

The invention according to claim 44 is the fading frequency determination apparatus according to claim 43, wherein the difference calculation for calculating a difference between inner product values for two inner product measurement intervals averaged by the inner product value averaging means. Means for judging the fading frequency using the difference calculated by the difference calculation means.

The invention according to claim 45 is the fading frequency determination device according to claim 38, wherein the inner product value calculating means is configured to calculate each of the two sections of the pilot channel for each of the multipaths used for RAKE combining. And a normalization means for normalizing the average value of the pilot symbols in each of the multipaths, and for each of the multipaths, the inner product value of the average value of the two pilot symbols normalized by the normalization means, Inner product value calculation executing means for calculating at least one inner product value; first inner product value averaging means for averaging the inner product value of each of the multipaths calculated by the inner product value calculating execution means for each inner product measurement interval; A second step of averaging the inner product value averaged by the first inner product value averaging means over a plurality of sections of the control channel for the measurement interval. Product value averaging means, wherein the determination means has determination execution means for determining a fading frequency using an inner product value for each inner product measurement interval averaged by the second inner product value averaging means. It is characterized by.

The invention according to claim 46 is the fading frequency determination device according to claim 45, wherein the difference between the inner product values for the two inner product measurement intervals averaged by the second inner product value averaging means is calculated. The image processing apparatus further includes a difference calculation unit, wherein the determination execution unit determines the fading frequency also using the difference calculated by the difference calculation unit.

47. The channel estimation method according to claim 47, wherein a step of generating a weighting coefficient for weighting and averaging pilot symbols time-multiplexed on a control channel parallel-multiplexed with a data channel; Weighting and averaging the pilot symbols using coefficients, and calculating a channel estimation value of the data symbols of the data channel.

The invention according to claim 48, comprising a fading frequency determination method, comprising the steps of: calculating an inner product value of a pilot symbol time-multiplexed on a control channel parallel-multiplexed with a data channel; and calculating a fading frequency based on the inner product value. And a step of determining

The invention according to claim 49 is a channel estimation method for calculating a channel estimation value of the data symbol using a pilot symbol in a channel in which the data symbol and the pilot symbol are time-multiplexed, wherein Dividing a data symbol into a plurality of data symbol sections, selecting a pilot symbol appropriate for obtaining a channel estimation value of the data symbol in each data symbol section, and generating a weighting coefficient for weighting and averaging the pilot symbols; And weighting and averaging the pilot symbols using the weighting coefficients, and calculating a channel estimation value of a data symbol in each data symbol section.

The invention according to claim 50 is a fading frequency determination method, wherein a step of calculating an inner product value of a pilot symbol in a channel in which a data symbol and a pilot symbol are time-multiplexed, and determining a fading frequency based on the inner product value And the step of performing.

The invention according to claim 51, is a channel estimation method for calculating a channel estimation value of a data symbol of the data channel using a pilot symbol of a pilot channel multiplexed in parallel with the data channel, wherein the data symbol of the data channel is Is divided into a plurality of data symbol intervals, a pilot symbol suitable for calculating a channel estimation value of the data symbol of each data symbol interval is selected, and a weighting coefficient for weighting and averaging the pilot symbols is generated. Weighting and averaging the pilot symbols using the weighting coefficients, and calculating a channel estimation value of a data symbol in each data symbol section.

An invention according to claim 52 is a fading frequency determination method, wherein the fading frequency is determined based on an inner product value of pilot symbols of a pilot channel multiplexed in parallel with a data channel.

The invention according to claim 53 is a demodulation device, wherein N channel estimation values are obtained by temporally weighting and averaging a pilot signal using N (N is a natural number of 2 or more) weight sequences. Channel estimating means, compensating means for compensating a data sequence using the respective channel estimation values, RAKE combining means for RAKE combining each of the compensated N data sequences, and N number of RAKE combining after the RAKE combining A reliability determining unit for selecting one data sequence having the highest reliability from the data sequence.

An invention according to claim 54 is a demodulation device, wherein for a data sequence of a predetermined number of frames, a pilot signal is temporally converted using N (N is a natural number of 2 or more) weight sequences. Channel estimation means for obtaining N channel estimation values by weighted averaging; compensation means for compensating a data sequence using each of the channel estimation values; and RAKE for RAKE combining each of the compensated N data sequences Combining means for selecting N ′ weight sequences with high reliability (N ′: natural number, N ′ <N) from the N data sequences after the RAKE combination, and selecting the most reliable weight sequence from the N data sequences A reliability determining means for selecting one data sequence having a high degree, the N 'weight sequences being selected at regular intervals, and the remaining data being determined until the next reliability determination is performed. About the series The channel estimation means obtains N 'channel estimation values by temporally weighting and averaging using the N' weight sequences, and the compensation means compensates the data sequence using the N 'channel estimation values. The RAKE combining means RAKE-combines each of the compensated N 'data sequences, and the reliability determining means selects one data sequence having the highest reliability from the N' data sequences. It is characterized by doing.

The invention according to claim 55 is the demodulation device according to claim 53 or 54, wherein the data sequence reliability determination unit performs error correction decoding of the RAKE-combined data sequence. A CRC bit extracting unit for extracting a CRC bit added to the data sequence, a CRC decoding unit for decoding a CRC of the data sequence, and detecting the presence or absence of a frame error based on a result of the CRC decoding. Frame error detecting means, frame error number counting means for counting the number of frame errors in a predetermined measurement time, and a demodulation using a highly reliable weight sequence and the weight sequence based on the frame error count result It has a weight sequence / data selection means for selecting a data sequence.

The invention according to claim 56 is the demodulation device according to claim 53 or 54, wherein the reliability determination unit of the data sequence performs error correction decoding of the data sequence after the RAKE combination. And likelihood information extracting means for extracting likelihood information calculated at the time of error correction decoding of each data sequence, and likelihood averaging means for averaging the extracted likelihood information for a predetermined measurement time. And a weight sequence / data selection means for selecting a highly reliable weight sequence based on the averaged likelihood information and a data sequence to be demodulated using the weight sequence.

The invention according to claim 57 is the demodulation device according to claim 53 or 54, wherein the data sequence reliability determination unit includes a power calculation unit that calculates power of each data sequence after the RAKE combination. Power averaging means for averaging the calculation result of the power for a predetermined measurement time, and a weight sequence having high reliability based on the averaged power and a data sequence demodulated using the weight sequence. And a weight sequence / data selecting means for selecting.

The invention according to claim 58 is the demodulation device according to claim 53 or 54, wherein the reliability determination unit for the data sequence is configured to perform an SN ratio (signal power to noise power) of each data sequence after the RAKE combining. Ratio), an SN ratio averaging means for averaging the calculation result of the SN ratio for a predetermined measurement time, and a weight sequence having high reliability based on the averaged SN ratio. And a weight sequence / data selection means for selecting a data sequence to be demodulated using the weight sequence.

The invention according to claim 59 is the demodulation device according to claim 53 or 54, wherein the data sequence reliability determination unit performs error correction decoding of the RAKE-combined data sequence. A CRC bit extracting unit for extracting a CRC bit added to the data sequence, a CRC decoding unit for decoding a CRC of the data sequence, and detecting the presence or absence of a frame error based on a result of the CRC decoding. Frame error detection means, frame error number counting means for counting the number of frame errors in a predetermined measurement time, and likelihood information extraction means for extracting likelihood information calculated at the time of error correction decoding of each data sequence; Means for averaging the extracted likelihood information for a predetermined measurement time, and a plurality of data systems Weighting sequence and data selecting means for selecting a highly reliable weight sequence based on the measured number of frame errors and the averaged likelihood information and a data sequence to be demodulated using the weight sequence. It is characterized.

The invention according to Claim 60 is the demodulation device according to Claim 53 or 54, wherein the reliability determination means for the data sequence performs error correction decoding for the data sequence after the RAKE combination. A CRC bit extracting unit for extracting a CRC bit added to the data sequence, a CRC decoding unit for decoding a CRC of the data sequence, and detecting the presence or absence of a frame error based on a result of the CRC decoding. Frame error detecting means, frame error number counting means for counting the number of frame errors in a predetermined measurement time, power calculating means for calculating power of each data series after the RAKE combination, and a calculation result of the power Power averaging means for averaging for a predetermined measurement time, the number of frame errors and the averaged power And having a weight-series data selection means for selecting a data series to be demodulated with a high weighting sequences reliability and the weight sequence based.

The invention according to Claim 61 is the demodulation device according to Claim 53 or 54, wherein the reliability determination means of the data sequence performs error correction decoding of the data sequence after the RAKE combination. A CRC bit extracting unit for extracting a CRC bit added to the data sequence, a CRC decoding unit for decoding a CRC of the data sequence, and detecting the presence or absence of a frame error based on a result of the CRC decoding. Frame error detecting means, frame error number counting means for counting the number of frame errors during a predetermined measurement time, SN ratio calculating means for calculating the SN ratio of each data series after the RAKE combination, Signal-to-noise averaging means for averaging the calculation result of the above for a predetermined measurement time; By using the higher weight sequence and the weight sequence reliability based on the SN ratio and having a weight-series data means for selecting a data series to be demodulated.

The invention according to claim 62 is a demodulation device, comprising: a channel estimation means for obtaining a plurality of channel estimation values by weighting and averaging a received pilot signal using a plurality of weight sequences; And demodulation means for outputting a plurality of demodulated data sequences using the channel estimation values, and reliability judging means for selecting one demodulated data sequence by judging the reliability of the plurality of demodulated data sequences. It is characterized by the following.

The invention according to claim 63 is the demodulation device according to claim 62, wherein the reliability determination means is configured to calculate the reliability of the plurality of weighted sequences based on the reliability determination result in the plurality of demodulated data sequences. And selecting means for selecting a predetermined number of weight sequences from the data, wherein the demodulation means performs demodulation using only the predetermined number of weight sequences when the predetermined number of weight sequences is selected.

The invention according to claim 64 is the demodulation device according to any one of claims 53 to 63, wherein the pilot signal is time-multiplexed on a control channel parallel-multiplexed with a data channel including the data sequence. It is characterized by having.

The invention according to claim 65 is the demodulation device according to any of claims 53 to 63, wherein the pilot signal is time-multiplexed on one channel together with the data sequence.

The invention according to claim 66 is the demodulation device according to claim 65, wherein the channel estimating means divides a data sequence in a slot of the channel into a plurality of data sequence sections, and A pilot signal suitable for calculating a channel estimation value of data is selected, and the pilot signal is weighted and averaged to calculate a channel estimation value of data in each data sequence section.

The invention according to claim 67 is the demodulation device according to any one of claims 53 to 63, wherein the pilot signal is included in a pilot channel parallel-multiplexed to a data channel including the data sequence. Features.

The invention according to claim 68 is the demodulation device according to claim 67, wherein the channel estimating means divides the data sequence into a plurality of data sequence sections, and calculates a channel estimation value of data in each data series section. Is selected, and the pilot signal is weighted and averaged to calculate a channel estimation value of data in each data sequence section.

An invention according to claim 69 is a demodulation method, wherein a pilot signal is temporally weighted and averaged using N (N is a natural number of 2 or more) weight sequences to obtain N channel estimation values. Compensating a data sequence using each of the channel estimation values; RAKE combining each of the N data sequences after the compensation; and most reliable from the N data sequences after the RAKE combination. And a reliability determination step of selecting one data sequence having a high degree of reliability.

An invention according to claim 70 is a demodulation method, wherein a pilot signal is temporally weighted using N (N is a natural number of 2 or more) weight sequences for a data sequence having a predetermined number of frames. Averaging to obtain N channel estimation values, compensating a data sequence using the respective channel estimation values, RAKE combining each of the compensated N data sequences; N ′ weight sequences with high reliability (N ′: natural number, N ′ <N) are selected from the N data sequences after synthesis, and one of the N data sequences having the highest reliability is selected. A reliability determination step of selecting a data sequence, the N ′ weight sequences are selected at regular intervals, and the channel is selected for the remaining data sequences during the period until the reliability determination is performed next. Push The step of performing time-weighted averaging using the N ′ weight sequences to obtain N ′ channel estimates, and the step of compensating comprises compensating the data sequence using the N ′ channel estimates. The RAKE combining step RAKE-combines each of the compensated N ′ data sequences, and the reliability determining step includes determining one data sequence having the highest reliability from the N ′ data sequences. It is characterized by selecting.

The invention according to claim 71 is the demodulation method according to claim 69 or 70, wherein the reliability determination step includes a step of performing error correction decoding of the data series after the RAKE combination, and Extracting the added CRC bits, decoding the CRC for the data sequence, detecting the presence or absence of a frame error from the result of the decoding of the CRC, The method includes the steps of counting the number of frame errors and selecting a highly reliable weight sequence based on the frame error count result and a data sequence to be demodulated using the weight sequence.

The invention according to claim 72 is the demodulation method according to claim 69 or 70, wherein the reliability determination step includes a step of performing error correction decoding of the data series after the RAKE combining, Extracting the likelihood information calculated at the time of error correction decoding; averaging the extracted likelihood information for a predetermined measurement time; and determining the degree of reliability based on the averaged likelihood information. Selecting a high weight sequence and a data sequence to be demodulated using the high weight sequence.

The invention according to claim 73 is the demodulation method according to claim 69 or 70, wherein the reliability determination step includes a step of calculating the power of each data series after the RAKE combining, and a step of calculating the power. Averaging the result for a predetermined measurement time, and selecting a highly reliable weight sequence based on the averaged power and a data sequence to be demodulated using the weight sequence. Features.

The invention according to claim 74 is the demodulation method according to claim 69 or 70, wherein the reliability determination step includes a step of calculating an SN ratio of each data series after the RAKE combination, and a step of calculating the SN ratio. Averaging the calculation result of the above for a predetermined measurement time, and selecting a weight sequence having high reliability based on the averaged S / N ratio and a data sequence to be demodulated using the weight sequence. It is characterized by having.

The invention according to claim 75 is the demodulation method according to claim 69 or 70, wherein the reliability determination step includes a step of performing error correction decoding of the data sequence after the RAKE combining, and Extracting the added CRC bits, decoding the CRC for the data sequence, detecting the presence or absence of a frame error from the result of the decoding of the CRC, Counting the number of frame errors, extracting the likelihood information calculated during error correction decoding of each data sequence, and averaging the extracted likelihood information for a predetermined measurement time, Based on the measured number of frame errors of the plurality of data series and the averaged likelihood information, Characterized by a step of selecting a data series to be demodulated with have weight sequence and the weight sequence.

The invention according to claim 76 is the demodulation method according to claim 69 or 70, wherein the reliability determination step includes a step of performing error correction decoding of the data series after the RAKE combination, and a step of adding the data series to the data series. Extracting a CRC bit that has been encoded, decoding the CRC for the data sequence, detecting the presence or absence of a frame error based on the result of the decoding of the CRC, and detecting the frame at a predetermined measurement time. Counting the number of errors, calculating the power of each received data sequence after the RAKE combining, averaging the calculation result of the power over a predetermined measurement time, Weight sequence with high reliability based on averaged power and data sequence demodulated using the weight sequence Characterized by a step of selecting.

The invention according to claim 77 is the demodulation method according to claim 69 or 70, wherein the reliability determination step includes a step of performing error correction decoding of the data series after the RAKE combination, and Extracting the added CRC bits, decoding the CRC for the data sequence, detecting the presence or absence of a frame error from the result of the decoding of the CRC, Counting the number of frame errors, calculating the SN ratio of each data sequence after the RAKE combination, averaging the calculation result of the SN ratio for a predetermined measurement time, And a weight sequence having high reliability based on the averaged SN ratio and data demodulated using the weight sequence. Characterized by a step of selecting a sequence.

The invention according to claim 78 is a demodulation method, comprising: a step of obtaining a plurality of channel estimation values by weighting and averaging a pilot signal using a plurality of weight sequences; A step of deriving a plurality of demodulated data sequences from the sequence and a step of selecting one output data sequence by determining the reliability of the plurality of demodulated data.

The invention according to claim 79 is the demodulation method according to claim 78, wherein a predetermined number of weight sequences are selected from the plurality of weight sequences based on a reliability determination result of the plurality of demodulated data sequences. After the selection, the demodulation is performed only by the selected weight sequence.

The invention according to claim 80 is the demodulation method according to any of claims 69 to 79, wherein the pilot signal is time-multiplexed on a control channel parallel-multiplexed with a data channel including the data sequence. It is characterized by having.

The invention according to claim 81 is the demodulation method according to any one of claims 69 to 79, characterized in that the pilot signal is time-multiplexed together with the data sequence on one channel.

The invention according to claim 82 is the demodulation method according to claim 81, wherein the step of estimating the channel includes dividing a data sequence in a slot of the channel into a plurality of data sequence sections, A pilot signal suitable for calculating a channel estimation value of data in a section is selected, and the pilot signal is weighted and averaged to calculate a channel estimation value of data in each data series section.

The invention according to claim 83 is the demodulation method according to any one of claims 69 to 79, wherein the pilot signal is included in a pilot channel parallel-multiplexed with a data channel including the data sequence. Features.

The invention according to claim 84 is the demodulation method according to claim 83, wherein the step of estimating the channel comprises dividing the data sequence into a plurality of data sequence sections, and A pilot signal suitable for calculating an estimated value is selected, and the pilot signal is weighted and averaged to calculate a channel estimated value of data in each data sequence section.

According to the above configuration, highly accurate channel estimation can be performed by weighting and averaging pilot symbols and calculating channel estimation values of data symbols of data channels in the parallel time multiplexing method.

Further, a data symbol in a slot is divided into a plurality of data symbol sections, a pilot symbol appropriate for calculating a channel estimation value of the data symbol in each data symbol section is selected, and the pilot symbol is weighted and averaged to obtain a data symbol. By calculating the channel estimation value of the data symbol in the section, highly accurate channel estimation can be performed.

Furthermore, the fading frequency can be determined based on the inner product value of the pilot symbols. Further, it is possible to realize optimal channel estimation with respect to a fading frequency with a simpler configuration.

Further, in the configuration of the present invention, from the weight sequence that is effective when the moving speed is low and the averaging time is increased to some extent, it is effective when the moving speed is high and the averaging time is reduced to some extent. By preparing a plurality of weight sequences up to such weight sequences and performing demodulation processing in parallel using all of the plurality of weight sequences at all times or at certain fixed time intervals, it is effective for various moving speeds. By directly determining and using a proper weight sequence from the reception quality, it is possible to improve communication quality, reduce transmission power, and increase capacity.

Use of weighting factors corresponding to various moving speeds by always performing channel estimation using multiple weighting factors and selecting data sequences and weighting factors with high reliability by reliability determination using received data sequences , And highly accurate channel estimation becomes possible.

Furthermore, it is possible to reduce the load on the system by periodically selecting a small number of weighting factors and performing channel estimation using only the selected weighting factors during a certain period.

Also, by directly determining and using weight sequences effective for various moving speeds from reception quality, it is possible to improve communication quality, reduce transmission power, and increase communication capacity.

According to the present invention, highly accurate channel estimation can be performed by weighting and averaging pilot symbols and calculating channel estimation values of data symbols of data channels in the parallel time multiplexing method.

Further, a data symbol in a slot is divided into a plurality of data symbol sections, a pilot symbol appropriate for calculating a channel estimation value of the data symbol in each data symbol section is selected, and the pilot symbol is weighted and averaged to obtain a data symbol. By calculating the channel estimation value of the data symbol in the section, highly accurate channel estimation can be performed.

Furthermore, the fading frequency can be determined based on the inner product value of the pilot symbols. Further, it is possible to realize optimal channel estimation with respect to a fading frequency with a simpler configuration.

If highly accurate channel estimation can be realized, the absolute synchronous detection can reduce the SNIR required to obtain a required reception quality (reception error rate), and as a result, the transmission power can be reduced. User capacity can be increased.

The determination result obtained by the fading frequency determination unit includes not only the setting of weighting factors in channel estimation, but also the switching of transmission power control operation and non-operation, the transmission diversity operation activation and non-operation, and the like. The transmission speed can be further improved by using the operation switching or parameter setting of various individual technologies whose performance (transmission characteristics) is affected by the moving speed of (1).

According to the present invention, it is possible to not only improve communication quality but also reduce transmission power and increase communication capacity by directly determining and using a weight sequence effective for moving speed from reception quality. More specifically, the following effects can be obtained.

(1) Since a weight sequence suitable for various moving speeds can be sequentially selected, highly accurate channel estimation can be performed. As a result, transmission power can be reduced, reception quality can be improved, and communication capacity can be increased.

(2) The system can be simplified by using only the selected sequence in the weight sequence except for the fixed time.

(3) By using CRC for data reliability determination, highly accurate reliability determination is possible.

(4) By using the likelihood obtained at the time of decoding the FEC at the time of determining the reliability of data, highly accurate reliability determination can be performed.

(5) By using the power or the SN ratio of the data sequence at the time of determining the reliability of data, it is possible to perform higher-speed and simpler reliability determination, thereby suppressing an increase in hardware scale.

Hereinafter, the best mode for carrying out the present invention will be described in detail with reference to the drawings.

(1st Embodiment)
FIG. 1 is a diagram illustrating an example of a frame configuration of a signal received by the demodulation device according to the first embodiment of the present invention. The demodulation device according to the present embodiment receives and demodulates a data channel and a control channel signal multiplexed in parallel with the data channel. Pilot symbols with a known transmission pattern (for example, with a known phase when the primary modulation is phase modulation) are time-multiplexed on the control channel (parallel time multiplexing method). Using the received signal (phase and amplitude) at the pilot symbol portion as a reference signal, the channel fluctuation of the data symbol of the data channel is estimated.

FIG. 2 is a diagram for explaining a method of channel estimation by the demodulation device according to the present embodiment. Channel estimation is performed using pilot symbols. Specifically, in a plurality of slots, an average of pilot symbols (complex fading envelope estimation values) ξ is obtained (by in-phase addition), and the average value ξ ′ is weighted as a weighting coefficient (coefficient used for weighted averaging) α _0. , by weighting averaged by alpha ₁ or the like carried out by calculating a channel estimate xi] ''.

In the example of FIG. 2, the channel estimation value ξ ″ (n) of the data symbol of the n-th slot is calculated by averaging the pilot symbol of the (n−2) -th pilot block (a set of pilot symbols of the (n−2) -th slot). Is calculated as follows using the channel estimation value ξ ′ (n + 3) of the (n + 3) th pilot block from the channel estimation value ξ ′ (n−2) obtained from

　異なるスロットに属する多くのパイロットシンボルを用いてチャネル推定を行うことにより高精度なチャネル推定を行うことができる。実際の移動伝搬環境においては、熱雑音（送信電力をできるだけ低減させるために、特にセル端では雑音の影響が大きい）、および他ユーザからの相互相関に起因する干渉信号が、自チャネルの希望波信号に加わり、さらに、フェージングによって受信信号の位相や振幅が時々刻々と変化するためにチャネル推定精度は劣化するからである。スロット単位で送信電力制御を行っている場合には、スロットが異なるパイロットシンボル間では電力が異なるが、この差に起因するチャネル推定誤差よりも、より多くのスロットのパイロットシンボルを用いることによる熱雑音、干渉信号の影響の低減効果の方が大きい。

By performing channel estimation using many pilot symbols belonging to different slots, highly accurate channel estimation can be performed. In an actual mobile propagation environment, thermal noise (in order to reduce transmission power as much as possible, especially at the cell edge, the effect of noise is large) and interference signals caused by cross-correlation from other users are generated by the desired signal of the own channel. This is because, in addition to the signal, the phase estimation and the amplitude of the received signal change every moment due to fading, so that the channel estimation accuracy deteriorates. When transmission power control is performed in slot units, the power differs between pilot symbols in different slots, but the thermal noise caused by using pilot symbols in more slots than the channel estimation error caused by this difference. The effect of reducing the influence of the interference signal is greater.

FIG. 3 is a block diagram showing a configuration example of the demodulation device according to the present embodiment. The demodulation device according to the present embodiment includes a matched filter for data channel 102, a delay unit 104, a matched filter for control channel 106, a channel estimation unit 120, a multiplication unit 108, and a RAKE (Rake) combining unit 110. Although the demodulation device according to the present embodiment is based on the CDMA (Code Division Multiple Access) system, the present invention uses other systems (for example, the TDMA (Time Division Multiple Access) system, the FDMA (Frequency Division Multiple Access) system) ) Can be applied to a demodulation device conforming to the above.

FIG. 4 is a block diagram illustrating a configuration example of the channel estimation unit according to the present embodiment. The channel estimation unit 120 according to the present embodiment includes a slot synchronization detection unit 122, a pilot symbol averaging unit 124, delay units 126, 128, and 130, multiplication units 132, 134, and 136, a weighting coefficient control unit 138, and an addition unit. 140 and a fading frequency determination unit 150. The channel estimation unit 120 can be realized as hardware, or can be realized as software using a DSP (Digital Signal Processor) or the like.

FIG. 5 is a block diagram illustrating a configuration example of the fading frequency determination unit according to the present embodiment. The fading frequency determination unit 150 according to the present embodiment includes a normalization unit 152, an inner product value calculation unit 154, a first averaging unit 156, a second averaging unit 158, and a determination unit 160.

Hereinafter, the operation of the demodulation device according to the present embodiment will be described with reference to FIGS. The data channel matched filter 102 despreads the received spread signal of the data channel using a spread code replica corresponding to the reception timing of each multipath of each user. The matched filter for control channel 106 despreads the received spread signal of the control channel using a spread code replica according to the reception timing of each user's multipath. The slot (pilot block) synchronization detector 122 of the channel estimator 120 detects a pilot symbol position in the control channel. From the timing information, pilot symbol averaging section 124 averages the reception channels of the pilot symbols in each pilot block to estimate a channel for each pilot block.

The estimated channel information in each pilot block is input to delay sections 126, 128, 130, etc., and their timings are aligned. Using the weighting factors generated by weighting factor control section 138, multiplication sections 132, 134, 136, etc. The addition section 140 calculates a channel estimation value by weighted averaging (weighted addition).

The channel estimation value of the data symbol of the n-th slot is, for example, as shown in FIG. 2, from the (n−K + 1) -th (K: natural number) pilot block to the (n + K) -th pilot block (K = 3 in the example of FIG. 2). It can be calculated using: Also, for example, in consideration of delay, the calculation can be performed using the (n−K + 1) th pilot block to the nth pilot block.

FIG. 6 is a diagram illustrating a calculation example of a channel estimation value. In the example of FIG. 6, the channel estimation value of the data symbol of the n-th slot is calculated using the (n-1) -th pilot block to the (n + 1) -th pilot block. Here, the ratio of the weighting coefficients may be, for example, α ₋₁ : α ₀ : α ₁ = 0.4: 1.0: 0.4. It is preferable that the value of the weighting factor is set larger for a pilot block closer (temporally closer) to a data symbol whose channel estimation value is to be calculated. This is because the propagation path changes every moment, and such a pilot block reflects the state of the propagation path when the n-th data symbol is transmitted. In the frame configuration of FIG. 6, since the position of the pilot block (pilot symbol) in the slot is earlier in time (as shown in FIG. 6, it is shifted to the left), the ratio of the weighting coefficient For example, it is considered that better channel estimation values can be obtained by setting α ₋₁ : α ₀ : α ₁ = 0.2: 1.0: 0.6. As described above, by determining the weighting coefficient according to the position of the pilot symbol in the slot, a highly accurate channel estimation value can be obtained.

2 and 6, the channel estimation value is calculated using all pilot symbols in the slot, but the channel estimation value may be calculated without using all pilot symbols in the slot. . In FIGS. 2 and 6, weighted averaging is performed after calculating the average value of pilot symbols in the pilot block. However, weighted averaging may be performed by providing a weighting coefficient for each pilot symbol. . When the number of pilot symbols in a pilot block is one, there is no need to calculate an average value.

2 and 6, the channel estimation value is common to all the data symbols in one slot. However, the data symbol in the slot is divided into a plurality of data symbol sections, and the channel estimation value is calculated for each data symbol section. Can be selected, and the pilot symbols can be weighted and averaged to calculate the channel estimation value of the data symbol in each data symbol section.

FIG. 7 is a diagram showing an example in which a data symbol in one slot is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. In the example of FIG. 7, for the data symbol sections (1), (2) and (3), the channel estimation value is calculated using the (n-1) th pilot block to the (n + 1) th pilot block, and the data symbol section (4 ), (5) and (6), the channel estimation value is calculated using the (n + 2) th pilot block from the nth pilot block. The calculation of the channel estimation values for the data symbol sections (1), (2), and (3) can be performed using the same weighting coefficients, or can be performed using different weighting coefficients. The same applies to data symbol sections (4), (5) and (6).

In the example of FIG. 7, the channel estimation value of the data symbol of the last data symbol section (1) of the (n-1) -th slot is calculated, and the data symbol of the first data symbol section (2) of the n-th slot is calculated. In the calculation of the channel estimation value, the same pilot symbol is selected, the pilot symbol is weighted and averaged, and the channel estimation value of the data symbol in each data symbol section is calculated.

FIGS. 8 to 10 are also diagrams showing examples in which a data symbol in one slot is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. In the examples of FIGS. 8 to 10, two symbols before and two symbols after the slot of the control channel are pilot symbols. In channel estimation, an average value for a fixed number of pilot symbols is calculated while sequentially moving the symbol position. In the examples of FIGS. 8 to 10, the pilot symbols are directly weighted and averaged without finding the average value of the pilot symbols for each pilot block.

In the example of FIG. 8, the number of pilot symbols used for weighted averaging is four, and a data symbol in one slot is divided into three sections. In the example of FIG. 9, four pilot symbols are used for weighted averaging, and a data symbol in one slot is divided into five sections. In the example of FIG. 10, eight pilot symbols are used for weighted averaging, and a data symbol in one slot is divided into three sections.

8 to 10, the calculation of the channel estimation value of the data symbol in the last data symbol section of the i-th (i: integer) slot and the channel of the data symbol in the first data symbol section of the (i + 1) -th slot In the calculation of the estimated value, the same pilot symbol is selected, the pilot symbol is weighted and averaged, and the channel estimated value of the data symbol in each data symbol section is calculated.

In the present embodiment, the weighting coefficients used for weighted averaging are changed according to the fading frequency. Fading frequency determination section 150 determines a fading frequency based on the average value of pilot symbols, and weighting coefficient control section 138 changes a weighting coefficient generated based on the determination result.

Fading frequency determination section 150 calculates an inner product value after normalizing the average value of pilot symbols in each of the two slots of the control channel.

FIGS. 11A and 11B are diagrams for explaining the concept of fading frequency determination. As shown in FIG. 11A, if the fading fluctuation is slow (the fading frequency is small), the correlation between the channel estimation values for each slot is large, and the inner product value is large. On the other hand, as shown in FIG. 11B, if the fading variation is fast (the fading frequency is large), the correlation between the channel estimation values for each slot is small, and the inner product value is small.

FIG. 12 is a diagram showing a result obtained by computer simulation of a measurement value (vertical axis) with respect to a measurement time (horizontal axis) using the fading frequency (fDTslot) as a parameter. In the example of FIG. 12, in order to determine whether or not the fading frequency is fast fading of 0.3 or more, the threshold value for the measured value is set to, for example, 0.3. It can be determined that the fading frequency is 3 or more.

The normalization unit 152 of the fading frequency determination unit 150 normalizes the average value of the pilot symbols in each of the two slots of the control channel, that is, the average value of the pilot symbols in the two pilot blocks. The inner product value calculator 154 calculates the inner product value of the average value of the two normalized pilot symbols.

The demodulation device according to the present embodiment is a demodulation device that performs RAKE combining, and performs the above-described normalization and inner product value calculation for each of the multipaths used for RAKE combining. The inner product value of each of the multipaths is averaged by the first averaging unit 156. When averaging is not performed over a plurality of paths, the first averaging unit 156 is unnecessary.

The average value calculated by the first averaging unit 156 is further averaged over a plurality of slots by the second averaging unit 158 (for example, the inner product values (1), (2), and (3) in FIG. 11A are averaged). Is changed). This reduces the effects of thermal noise. When averaging is not performed over a plurality of slots, second averaging section 158 is unnecessary.

The threshold value determining unit 160 determines the fading frequency by comparing the average value calculated by the second averaging unit 158 with the threshold value. More specifically, a threshold is determined in a plurality of steps based on a preset threshold to determine which of the plurality of regions the fading frequency corresponds to. In the present embodiment, the determination of the fading frequency is performed based on the threshold value.

In the present embodiment, the fading frequency is determined by calculating the inner product of the average values of the pilot symbols of the two pilot blocks. However, the two pilot blocks that take the inner product are, for example, pilot blocks of adjacent slots ( For example, pilot blocks (1) and (2) in FIG. 11A may be used, or pilot blocks of every other slot (for example, pilot blocks (1) and (3) in FIG. 11A) may be used. Further, the fading frequency may be determined by using an inner product of a certain pilot symbol and another pilot symbol without using a pilot block.

Further, when the inner product value (average value) of the pilot symbol (average value) (for example, the output of the second averaging unit 158 in FIG. 5) is larger than a certain value, the farther interval of the control channel Normalizing, calculating the inner product value, averaging the inner product value of each of the multipaths, and averaging the inner product value over the plurality of slots, for the average value of the pilot symbols in each of the two slots. The fading frequency can be determined by comparing the averaged inner product value with a threshold value corresponding to the farther interval.

As can be seen from the graph of the inner product value using the fading frequency as a parameter in FIG. 12, the difference in the inner product value due to the difference in frequency is relatively large (the resolution is high) at a higher fading frequency, so that the fading frequency is easily thresholded. While the value can be determined, the fading frequency determination tends to be difficult at lower fading frequencies because the difference in the inner product value is relatively small (resolution is low).

Here, the resolution at a lower fading frequency can be increased by making the interval (slot of the inner product measurement) of the slot including the pilot symbol used for calculating the inner product value longer. Therefore, first, an inner product value having a low resolution (that is, using a pilot symbol of a slot with a short interval) is obtained, and an inner product value larger than a certain value (that is, a frequency lower than a certain fading frequency) is obtained. Furthermore, by using the inner product value having a higher resolution (that is, using pilot symbols of slots with longer intervals) as the determination value of the fading frequency, the accuracy of a higher frequency range from a higher fading frequency to a lower fading frequency can be improved. It is possible to make a high determination.

For example, the inner product value (the average value) of pilot symbols (the average value) of pilot symbols in adjacent slots (the inner product measurement interval = 1 slot interval) (for example, the output of the second averaging unit 158 in FIG. 5) is a certain fixed value. If the value corresponds to a fading frequency equal to or lower than the frequency, it is possible to determine the fading frequency with higher resolution by determining the inner product value of pilot symbols at two-slot intervals further apart by one threshold. is there.

Further, when the inner product value of the two-slot interval is a value corresponding to a lower fading frequency equal to or lower than a certain fixed frequency, the fading frequency determination is performed by using the inner product value of the pilot symbol of the three-slot interval further away by one slot. It is possible to increase the resolution by gradually increasing the inner product measurement interval (the reason for changing the inner product measurement interval from a narrow interval to a wide interval is that for a given inner product measurement interval, This is because the frequency that can be determined becomes lower as the interval is increased.)

Note that the calculation of the inner product value at different inner product measurement intervals can be performed in parallel, and by calculating in parallel, even when performing the above-described stepwise determination, the determination can be made in a short time. The result can be obtained.

It is also possible to calculate two or more inner product values by changing the inner product measurement interval and determine the fading frequency using the inner product values.

FIGS. 13A and 13B are block diagrams illustrating another configuration example of the fading frequency determination unit 150 according to the present embodiment. 13A and 13B includes a normalization unit 162, delay units 163-1 and 163-2, inner product value calculation units 164-1 and 164-2, and first averaging units 166-1 and 166. -2, a second averaging unit 168-1, 168-2, a difference calculation unit 169, and a determination unit 170.

13A and 13B, the inner product value calculation unit 164-1 calculates the inner product value using the inner product measurement interval as one slot length, and the inner product value calculation unit 164-2 calculates the inner product value as two slot lengths (1 slot). (Skipping) calculating the inner product value.

The inner product value at each inner product measurement interval was averaged over a plurality of paths by the first averaging units 166-1 and 166-2, and averaged over a plurality of slots by the second averaging units 168-1 and 168-2. Thereafter, the difference calculator 169 calculates the difference between the inner product values at two inner product measurement intervals (the difference between the inner product value at one slot interval and the inner product value at two slot intervals). Then, determination section 170 determines the fading frequency using the inner product value at one slot interval, the inner product value at two slot intervals, and the difference therebetween.

In the examples of FIGS. 13A and 13B, averaging is performed over a plurality of paths and averaging over a plurality of slots. However, one or both of them may not be performed.

FIG. 14 is a diagram illustrating an example of determining a fading frequency. In the example of FIG. 14, the point P ₁ (the point where the inner product value and the difference (absolute value) at the interval of two slots match first) and the point P ₂ (the inner product value and the difference at the interval of one slot match first) The fading frequency is determined using the point P ₃ (the point where the inner product value at one slot interval and the inner product value at the two slot interval first match). I.e., less than the fading frequency at the point P _1, or less than the fading frequency at the point P ₂ in the fading frequency or at the point P _1, or less than the fading frequency at the point P ₃ in the fading frequency or at the point P _2, the fading at the point P ₃ The fading frequency is determined in four ways, that is, higher than the frequency.

れ ば If the determination is made in this way, the threshold does not have to be set. Further, it is possible to make a more detailed determination than when one inner product value is calculated without changing the inner product measurement interval. If more inner product values are calculated by changing the inner product measurement interval, more detailed judgment can be made.

The fading frequency may be determined using only a plurality of inner product values without calculating the difference. In that case, it will be determined using only the point P ₃ in the example of FIG. 14.

Based on the fading frequency determined in this way, the weighting coefficient control unit 138 changes the weighting coefficient. Considering the example of FIG. 6, when the fading frequency is high, the weighting coefficient of the pilot block closer (temporally close) to the data symbol for which the channel estimation value is to be calculated is larger than when the fading frequency is low. Enlarge. This is because, when the fading frequency is large, channel fluctuation of a data symbol whose channel estimation value is to be calculated is significantly different from channel fluctuation of a pilot block far (temporally far) from the data symbol. For example, the ratio of the weighting factor when the fading frequency is low is α ₋₁ : α ₀ : α ₁ = 0.2: 1.0: 0.6, and the ratio of the weighting factor when the fading frequency is high is α. ₋₁ : α ₀ : α ₁ = 0.05: 1.0: 0.5 (the pilot block of the n-th slot, the pilot block of the (n + 1) -th slot, and the pilot block of the (n−1) -th slot in this order. It is considered to be close to the data symbol for which the channel estimate is to be calculated).

In the present embodiment, the weighting coefficient used for weighted averaging is changed according to the fading frequency, but a fixed weighting coefficient may be used.

遅 延 Using the channel estimation value obtained in this way (the output of the adding unit 140), the delay unit 104 compensates for the channel fluctuation (fading fluctuation) of the despread data symbol whose timing has been planned. Specifically, the channel fluctuation is compensated by multiplying the despread data symbol by the complex conjugate of the channel estimation value. Then, the compensated signal is combined in phase by RAKE combining means 110.

In the present embodiment, a case has been described where the transmission rate of the data channel and the transmission rate of the control channel are the same, but the two transmission rates may be different.

FIG. 15 is a diagram illustrating an example in which the transmission rate of the data channel is different from the transmission rate of the control channel. In the example of FIG. 15, the transmission rate of the control channel is に of the transmission rate of the data channel. Even when the transmission rates are different as described above, it is possible to calculate a channel estimation value using pilot symbols.

(2nd Embodiment)
FIG. 16 is a diagram illustrating an example of a frame configuration of a signal received by the demodulation device according to the second embodiment of the present invention. The demodulation device according to the present embodiment receives and demodulates a signal of a channel (time multiplex method) in which data symbols and pilot symbols are time-multiplexed. Using the received signal (phase and amplitude) at the pilot symbol portion as a reference signal, the channel fluctuation of the data symbol is estimated. Pilot symbols are inserted at regular intervals between data symbols. The channel estimation method by the demodulation device according to the present embodiment is the same as the channel estimation method by the demodulation device according to the first embodiment of the present invention.

FIG. 17 is a block diagram illustrating a configuration example of the demodulation device according to the present embodiment. The demodulation device according to the present embodiment includes a matched filter 202, a delay unit 204, a channel estimation unit 220, a multiplication unit 208, and a RAKE (Rake) synthesis unit 210. Although the demodulation device according to the present embodiment is also based on the CDMA system, the present invention can be applied to a demodulation device based on another system (for example, the TDMA system or the FDMA system). The demodulation device according to the present embodiment performs multiple access transmission by spreading a wideband signal with a spreading code faster than the information rate.

構成 The configuration example of the channel estimation unit 220 according to the present embodiment is the same as the configuration example of the channel estimation unit 120 according to the first embodiment of the present invention shown in FIG. The slot synchronization detecting section 122 detects a pilot symbol position in a channel in which data symbols and pilot symbols are time-multiplexed. The configuration example of the fading frequency determination unit according to the present embodiment is also the same as the configuration example of the fading frequency determination unit 150 according to the first embodiment of the present invention illustrated in FIG. 5 (the configuration illustrated in FIGS. 13A and 13B). It is also possible to do).

動作 The operation of the demodulation device according to the present embodiment is basically the same as the operation of the demodulation device according to the first embodiment of the present invention.

FIG. 18 is a diagram showing an example in which a data symbol in one slot is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. In the example of FIG. 18, for the data symbol sections (1) and (2), channel estimation values are calculated using the (n-1) th pilot block to the (n + 1) th pilot block, and the data symbol sections (3) and (4) are calculated. For ()), the channel estimation value is calculated using the (n + 2) th pilot block from the nth pilot block. The calculation of the channel estimation value for the data symbol sections (1) and (2) can be performed using the same weighting coefficient, or can be performed using different weighting coefficients. The same applies to data symbol sections (3) and (4).

Also, in the example of FIG. 18, the channel estimation value of the data symbol of the last data symbol section (1) of the (n-1) -th slot is calculated, and the data symbol of the first data symbol section (2) of the n-th slot is calculated. In the calculation of the channel estimation value, the same pilot symbol is selected, the pilot symbol is weighted and averaged, and the channel estimation value of the data symbol in each data symbol section is calculated.

FIGS. 19 to 21 are also diagrams showing examples in which a data symbol in one slot is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. In the examples of FIGS. 19 to 21, two symbols before and two symbols after the slot of the control channel are pilot symbols. In channel estimation, an average value for a fixed number of pilot symbols is calculated while sequentially moving the symbol position. In the examples of FIGS. 19 to 21, the pilot symbols are directly weighted and averaged without finding the average value of the pilot symbols for each pilot block.

In the example of FIG. 19, four pilot symbols are used for weighted averaging, and the data symbols in one slot are divided into three sections. In the example of FIG. 20, four pilot symbols are used for weighted averaging, and a data symbol in one slot is divided into five sections. In the example of FIG. 21, eight pilot symbols are used for weighted averaging, and a data symbol in one slot is divided into three sections.

In the examples of FIGS. 19 to 21, the calculation of the channel estimation value of the data symbol in the last data symbol section of the i-th (i: integer) slot and the channel of the data symbol in the first data symbol section of the (i + 1) -th slot In the calculation of the estimated value, the same pilot symbol is selected, the pilot symbol is weighted and averaged, and the channel estimated value of the data symbol in each data symbol section is calculated.

も Also in the present embodiment, the weighting coefficient used for weighted averaging is changed according to the fading frequency. However, a fixed weighting coefficient can be used.

Using the channel estimation value obtained by the channel estimation unit 220, the delay unit 204 compensates for the channel fluctuation (fading fluctuation) of the data symbol after despreading, which is timed. Specifically, the channel fluctuation is compensated by multiplying the despread data symbol by the complex conjugate of the channel estimation value. Then, the compensated signal is combined in phase by RAKE combining means 210.

に お い て In the present embodiment, the transmission rates of the data symbols and the pilot symbols in the channel are the same, but the transmission rates of the data symbols and the pilot symbols in the channel may be different.

(Third embodiment)
The above concept can be applied to the parallel system.

FIG. 22 is a diagram illustrating an example of a frame configuration of a signal received by the demodulation device according to the third embodiment of the present invention. The demodulation device according to the present embodiment receives and demodulates a data channel and a pilot channel signal (parallel system) multiplexed in parallel with the data channel. Using the received signal (phase and amplitude) of the pilot symbol of the pilot channel as a reference signal, the channel fluctuation of the data symbol of the data channel is estimated. In the parallel method, the pilot symbol is not transmitted and received using a part of the slot as in the parallel time multiplexing method and the time multiplexing method, but the pilot symbol is continuously transmitted and received, so the concept of the slot is not so important. . Therefore, the slots are not specifically shown in FIG.

The channel estimation method by the demodulation device according to the present embodiment is basically the same as the channel estimation method by the demodulation device according to the first embodiment and the second embodiment of the present invention. This will be explained.

FIG. 23 is a block diagram illustrating a configuration example of the demodulation device according to the present embodiment. The demodulation device according to the present embodiment includes a matched filter for data channel 302, a delay unit 304, a matched filter for pilot channel 306, a channel estimation unit 320, a multiplication unit 308, and a RAKE (Rake) combining unit 310. Although the demodulation device according to the present embodiment is also based on the CDMA system, the present invention can be applied to a demodulation device based on another system (for example, the TDMA system or the FDMA system).

FIG. 24 is a block diagram illustrating a configuration example of a channel estimation unit according to the present embodiment. The channel estimation unit 320 according to the present embodiment includes a pilot symbol averaging unit 324, delay units 326, 328, 330, etc., multiplication units 332, 334, 336, etc., a weighting coefficient control unit 338, an addition unit 340, and a fading frequency determination. A section 350 is provided. A configuration example of the fading frequency determination unit (fading frequency determination unit 350) according to the present embodiment is the same as the configuration example of the fading frequency determination unit 150 according to the first embodiment of the present invention illustrated in FIG. 5 (FIG. 13A). And the configuration as shown in FIG. 13B).

動作 The operation of the demodulation device according to this embodiment is basically the same as the operation of the demodulation device according to the first and second embodiments of the present invention.

FIG. 25 is a diagram illustrating an example in which a data symbol of a data channel is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. In the example of FIG. 25, a data symbol is divided into intervals of three symbols, and a channel estimation value is calculated using a temporally corresponding pilot symbol interval (three-symbol configuration) and pilot symbol intervals before and after the pilot symbol interval. . More specifically, a channel estimation value ξ ′ (0) obtained from an average of three symbols in a temporally corresponding pilot symbol section, and a channel estimation value か ら ′ obtained from an average in pilot symbols sections before and after the same. (−1) and ξ ′ (1) are weighted and averaged by α ₀ , α ₋₁ and α ₁ to calculate a channel estimation value ξ ″.

FIGS. 26 and 27 are also diagrams showing examples in which a data symbol of a data channel is divided into a plurality of data symbol sections (intervals of one symbol) and a channel estimation value is calculated for each data symbol section. In channel estimation, an average value for a fixed number of pilot symbols is calculated while sequentially moving the symbol position. In the examples of FIGS. 26 and 27, the pilot symbols are directly weighted and averaged, instead of calculating the average value of the pilot symbols and performing weighted averaging as in the example of FIG.

In the example of FIG. 26, there are four pilot symbols used for weighted averaging, and the pilot symbols used for weighted averaging are changed for each data symbol. In the example of FIG. 27, the number of pilot symbols used for weighted averaging is four, and the pilot symbol used for weighted averaging is changed for every two data symbols.

も Also in the present embodiment, the weighting coefficient used for weighted averaging is changed according to the fading frequency. However, a fixed weighting coefficient can be used.

FIGS. 28A and 28B are diagrams for explaining the concept of fading frequency determination. The fading frequency determination method in the present embodiment is basically the same as the fading frequency determination method in the first and second embodiments. In the first and second embodiments, the average value of pilot symbols in each of two slots is used. In the present embodiment, the average value of pilot symbols in each of two sections of a pilot channel is used. . The two sections may be discontinuous or continuous (discontinuous in FIG. 28). Further, one pilot symbol may be included in one section, or two or more pilot symbols may be included.

Also in the present embodiment, similarly to the first embodiment and the second embodiment, when the calculated inner product value is larger than a certain value, the interval (inner product) of the section including the pilot symbol used for calculating the inner product value is included. The inner product value can be calculated by increasing (far) the measurement interval). Further, two or more inner product values are calculated by changing the inner product measurement interval, and the fading frequency can be determined using the inner product values.

Using the channel estimation value obtained by the channel estimation unit 320, the delay unit 304 compensates for channel fluctuation (fading fluctuation) of the data symbol after despreading, which is timed. Specifically, the channel fluctuation is compensated by multiplying the despread data symbol by the complex conjugate of the channel estimation value. Then, the compensated signal is combined in phase by RAKE combining means 310.

も Also in this embodiment, similarly to the first embodiment, the transmission rate of the data channel and the transmission rate of the pilot channel can be different.

(Fourth embodiment)
First, a method of averaging pilot signals in channel estimation used in the fourth to tenth embodiments of the present invention will be described with reference to FIG. As shown in FIG. 29, a pilot symbol having a known transmission pattern (phase when primary modulation is phase modulation) is transmitted from a communication partner station together with information data symbols. At this time, the pilot symbols may be transmitted not only continuously as shown in FIG. 29 but also intermittently. That is, the pilot symbol insertion method may be a parallel time multiplexing method (FIG. 1), a time multiplexing method (FIG. 16), or a parallel method (FIG. 22).

(4) In order to estimate the phase variation of the propagation path and compensate the communication channel (phase correction), a channel estimation value is obtained by averaging pilot signals in sections before and after received data to be phase corrected. In the example of FIG. 29, in order to obtain the channel vector of the N-th symbol of the communication channel, the pilot signals before and after are divided into a weight sequence a (where a = ｛a (k) | k =. 1,...｝).

However, when averaging using a weight sequence, averaging using a weight sequence is performed after performing simple averaging on an arbitrary block of a chip unit or more.

In the fourth to tenth embodiments described in detail below, in a demodulation device conforming to the CDMA system by direct spreading, a pilot signal is weighted and averaged using a plurality of predetermined weight sequences to obtain a channel estimation value. Then, the received data is demodulated using the obtained channel estimation value, and the reliability of the plurality of demodulated data is determined to select one output data having the highest quality.

Also, for a certain period, it is possible to select some weight sequences based on the reliability determination result of the demodulated data sequence. In this case, thereafter, demodulation is performed only with the selected weight sequence.

(Configuration of Fourth Embodiment)
FIG. 30A and FIG. 30B are block diagrams showing the fourth embodiment. In this figure, 1 is a despreading unit, 2 (including 2-1 to 2-N) is a channel estimation unit, 3 (including 3-1 to 3-N) is a multiplier, and 4 (4-1 to 4-4). -N) is a RAKE combiner, 5 (including 5-1 to 5-N) is an FEC decoder, 6 (including 6-1 to 6-N) is a CRC decoder, and 7 (7-1 to 5-N). 7-N), a frame error number calculation section, 8A a reliability comparison section, 9 a reliability determination section, and 10 a first changeover switch.

(Operation of Fourth Embodiment)
Next, the operation of the CDMA demodulator of the fourth embodiment shown in FIGS. 30A and 30B will be described.

First, a received spread signal is input to the despreading unit 1, and the received spread data sequence is despread using a spreading code replica corresponding to the multipath timing.

The channel estimator 2 prepares N (N ≧ 2) weight sequences for averaging the pilot signals, and averages the pilot signals in parallel with the respective weight sequences to obtain a channel estimation value.

The multiplication unit 3 corrects the phase by multiplying the data sequence of the despread communication channel by the complex conjugate of each channel estimation value.

Next, in the RAKE combining section 4, the signals after the phase correction are in-phase combined in all the fingers, and input to the reliability determining section 9.

In the reliability judgment unit 9, first, the FEC decoding unit 5 decodes the error correction code, and outputs N pieces of decoded data of weighting factors # 1 to #N.

The CRC decoding unit 6 performs CRC decoding using the CRC bits extracted from the decoded data sequence, determines whether there is a frame error, and inputs the determination result to the frame error number calculation unit 7.

(4) The frame error number calculation unit 7 counts the number of frame errors existing between the predetermined number of frames, and inputs the count number to the reliability comparison unit 8.

The reliability comparison / determination unit 8A selects the data sequence of the system with the least number of frame errors from the N frame error information, and outputs the data by switching the first switch 10 to the desired system.

(Effect of Fourth Embodiment)
As described above, according to the fourth embodiment, channel estimation is always performed using a plurality of weighting factors, and data having high reliability is selected by reliability determination using a received data sequence. Weight coefficients corresponding to various moving speeds can be used at the same time, and highly accurate channel estimation becomes possible. Further, by selecting a weight sequence having a small number of frame errors using the CRC decoding result, it is possible to make a determination to reduce the frame error rate.

(Modification of Fourth Embodiment)
In the above description, the processes from channel estimation to CRC decoding using the N weight sequences are always performed. However, the following modifications can reduce the load on the system.

変 形 A modification of the fourth embodiment is shown in FIGS. 31A and 31B.

に お い て In FIGS. 31A and 31B, the same parts as those in the fourth embodiment shown in FIGS. 30A and 30B are denoted by the same reference numerals. Reference numeral 11 denotes a second changeover switch.

毎 At every fixed period, for the data series of a predetermined number of frames, all the second changeover switches 11 are turned ON, and the operation of the fourth embodiment is performed by N systems. In addition, the reliability determination unit 9 selects N ′ weight sequences (here, N ′: natural number, 1 ≦ N ′ <N) having a high reliability in the number of frames. After the reliability determination, for the remaining data series until the reliability determination is performed again at the time interval, only the switch of the selected weight series is turned on, and the switches of the other weight series are turned on. The operation becomes OFF, and the same operation as in the fourth embodiment is performed in the N ′ system using the selected N ′ weight sequences.

FIGS. 31A and 31B show an example in which two systems of weight sequence # 1 and weight sequence # 2 are selected (N ′ = 2), and only two systems are in operation.

(Fifth embodiment)
(Configuration of Fifth Embodiment)
FIG. 32 is a block diagram illustrating a reliability determination unit according to the fifth embodiment. The function blocks other than the reliability determination unit conform to the fourth embodiment and are omitted. The same parts as those in the fourth embodiment shown in FIGS. 30A and 30B are denoted by the same reference numerals. Reference numeral 12 (12-1 to 12-N) denotes a likelihood averaging unit.

(Operation of Fifth Embodiment)
Next, the operation of the reliability determination unit of the fifth embodiment will be described. The operation of the other functional blocks conforms to the fourth embodiment, and is omitted.

The RAKE-combined signal is input to the FEC decoder 5. The FEC decoding unit 5 decodes the error correction code, outputs decoded data of the weight sequences # 1 to #N, and inputs likelihood information calculated at the time of error correction to the likelihood averaging unit 12. .

The likelihood averaging unit 12 averages the inputted likelihood with a predetermined number of frames and Y frames (here, Y: natural number, Y ≧ 1), and inputs the average to the reliability comparison unit 8. The reliability comparing unit 8 selects a data series having the highest reliability as information output from the N pieces of likelihood information.

(Effect of Fifth Embodiment)
As described above, according to the fifth embodiment, by using the likelihood information calculated at the time of error correction decoding for reliability determination, it is possible to perform determination reflecting communication quality (such as a bit error rate). It becomes possible.

(Modification 1 of Fifth Embodiment)
In the above description, the processes from channel estimation to CRC decoding using the N weight sequences are always performed. However, the following modifications can reduce the load on the system.

変 形 A modification of the fifth embodiment can be configured by replacing the reliability determination unit of the modification of the fourth embodiment shown in FIGS. 31A and 31B with the fifth embodiment shown in FIG.

毎 At every fixed period, for the data series of a predetermined number of frames, all the second changeover switches 11 are turned ON, and the operation of the fourth embodiment is performed by N systems. In addition, the reliability determination unit selects N ′ weight sequences (here, N ′: natural number, 1 ≦ N ′ <N) having a high reliability in the number of frames. After the reliability determination, for the remaining data series until the reliability determination is performed again at the above time interval, the second changeover switch 11 turns ON only the selected weight series switch, and switches the other weight series switches. Is OFF, and the same operation as that of the fourth embodiment is performed in the N ′ system using the selected N ′ weight sequences.

(Modification 2 of Fifth Embodiment)
In the above description, the likelihood averaging unit 12 calculates the likelihood by a predetermined number of frames and Y frames (where Y is a natural number, Y ≧ 1), and also performs a weighted average or a Y-frame. , A method of selecting the minimum value, and a method of selecting the maximum value.

(Sixth embodiment)
(Configuration of Sixth Embodiment)
FIG. 33 is a block diagram illustrating a reliability determination unit according to the sixth embodiment. The function blocks other than the reliability determination unit conform to the fourth embodiment and are omitted. The same parts as those in the fourth embodiment shown in FIGS. 30A and 30B are denoted by the same reference numerals. Reference numeral 13 (including 13-1 to 13-N) denotes a power calculator.

(Operation of Sixth Embodiment)
Next, the operation of the reliability determination unit of the sixth embodiment will be described.

動作 The operation of the other functional blocks conforms to the fourth embodiment and will not be described.

The RAKE-combined signal is input to the power calculator 13. The power calculation unit 13 calculates the power of the N systems after the RAKE combining, and averages the power over a predetermined period.

The averaged power calculation value is input to the reliability comparison unit. The reliability comparison / determination unit 8 selects a data sequence having the highest reliability from the power calculation values of the N systems and inputs the data sequence to the FEC decoding unit 5. The error correction decoding is performed in the FEC decoding unit 5 and the information is output as an information output.

(Effect of Sixth Embodiment)
As described above, according to the sixth embodiment, by using the received power after the RAKE combination for reliability determination, it is possible to make a determination to increase the received power. Frame error rate) and the reliability can be determined without performing FEC decoding, so that the load on the system can be reduced.

(Modification of the sixth embodiment)
In the above description, the processes from channel estimation to power calculation using the N weight sequences are always performed. However, the following modifications can reduce the load on the system.

The modification of the sixth embodiment can be configured by replacing the reliability determination unit of the modification of the fourth embodiment shown in FIGS. 31A and 31B with the sixth embodiment shown in FIG.

毎 At every fixed period, for the data series of a predetermined number of frames, all the second changeover switches 11 are turned ON, and the operation of the fourth embodiment is performed by N systems. In addition, the reliability determination unit selects N ′ weight sequences (here, N ′: natural number, 1 ≦ N ′ <N) having a high reliability in the number of frames. After the reliability determination, for the remaining data series until the reliability determination is performed again at the above time interval, the second changeover switch 11 turns ON only the selected weight series switch, and switches the other weight series switches. Is OFF, and the same operation as that of the fourth embodiment is performed in the N ′ system using the selected N ′ weight sequences.

(Seventh embodiment)
(Configuration of Seventh Embodiment)
FIG. 34 is a block diagram illustrating a reliability determination unit according to the seventh embodiment. The function blocks other than the reliability determination unit conform to the fourth embodiment and are omitted. The same parts as those in the fourth embodiment shown in FIGS. 30A and 30B are denoted by the same reference numerals. Reference numeral 14 (including 14-1 to 14-N) denotes an SN ratio calculation unit.

(Operation of Seventh Embodiment)
Next, the operation of the reliability determination unit of the seventh embodiment will be described. The operation of the other functional blocks conforms to the fourth embodiment, and is omitted.

The RAKE-combined signal is input to the SN ratio calculator 14. The S / N ratio calculator 14 calculates the S / N ratios of the N systems after the RAKE combining, and averages them over a predetermined period. The averaged SN ratio calculation value is input to the reliability comparison unit 8.

The reliability comparing section 8 selects a data series having the highest reliability from the calculated SN ratios of the N systems and inputs the data series to the FEC decoding section 5. The error correction decoding is performed in the FEC decoding unit 5 and the information is output as an information output.

(Effect of Seventh Embodiment)
As described above, according to the seventh embodiment, by using the SN ratio after the RAKE combination for reliability determination, it is possible to make a determination to increase the SN ratio, so that the communication quality ( Frame error rate) can be improved, and the reliability can be determined before performing error correction decoding. Therefore, the load on the system can be reduced.

(Modification of Seventh Embodiment)
In the above description, the processes from channel estimation using the N weight sequences to SN ratio calculation are always performed. However, by modifying as follows, the load on the system can be reduced.

変 形 A modification of the seventh embodiment can be configured by replacing the reliability determination unit of the modification of the fourth embodiment shown in FIGS. 31A and 31B with the seventh embodiment shown in FIG.

For every predetermined period, for the data series of a predetermined number of frames, the second changeover switches 11 are all turned on, and the operation of the fourth embodiment is performed by N systems. In addition, the reliability determination unit selects N ′ weight sequences (here, N ′: natural number, 1 ≦ N ′ <N) having a high reliability in the number of frames. After the reliability determination, for the remaining data series until the reliability determination is performed again at the above time interval, the second changeover switch 11 turns ON only the selected weight series switch, and switches the other weight series switches. Is OFF, and the same operation as that of the fourth embodiment is performed in the N ′ system using the selected N ′ weight sequences.

(Eighth embodiment)
(Configuration of Eighth Embodiment)
FIGS. 35A and 35B are block diagrams illustrating a reliability determination unit according to the eighth embodiment. The function blocks other than the reliability determination unit conform to the fourth embodiment and are omitted. The same parts as those in the fourth embodiment of FIGS. 30A and 30B are denoted by the same reference numerals.

(Operation of Eighth Embodiment)
Next, the operation of the reliability determination unit of the eighth embodiment will be described. The operation of the other functional blocks conforms to the fourth embodiment, and is omitted.

The RAKE-combined signal is input to the FEC decoder 5. The FEC decoding unit 5 decodes the error correction code, outputs decoded data of weight coefficients # 1 to #N, and inputs likelihood information calculated at the time of error correction to the likelihood averaging unit 12. .

The likelihood averaging unit 12 averages the inputted likelihood by a predetermined number of frames, Y1 frames (here, Y1: natural number, Y1 ≧ 1), and inputs the average to the reliability comparison determination unit 8.

On the other hand, the CRC decoding unit 6 performs CRC decoding using the CRC bits extracted from the data sequence decoded by the FEC decoding unit 5, determines the presence or absence of a frame error, and returns the determination result to the frame error number calculation unit. Enter 7

The frame error number calculation unit 7 counts the number of frame errors existing in a predetermined Y2 frame (here, Y2: natural number, Y2 ≧ 1), and inputs the count number to the reliability comparison unit 8. .

In the reliability comparing section 8, from among the systems having the least number of frame errors than the N sets of frame error information output from the frame error number calculating section 7, the system having the least likelihood information input from the likelihood averaging section 12 is selected. A data sequence with high reliability is selected as the information output.

(Effects of Eighth Embodiment)
As described above, according to the eighth embodiment, the likelihood information calculated at the time of error correction decoding is used for the reliability determination together with the number of frame errors counted from the CRC decoding result, so that the mutual determination can be performed. By combining the factors, strict reliability determination can be performed.

(Modification 1 of the eighth embodiment)
In the above description, the processes from channel estimation to CRC decoding using the N weight sequences are always performed. However, the following modifications can reduce the load on the system.

The modification of the eighth embodiment is configured by replacing the reliability determination unit of the modification of the fourth embodiment shown in FIGS. 31A and 31B with the eighth embodiment shown in FIGS. 35A and 35B. .

{Circle around (2)} At every fixed period, all the second changeover switches 11 are turned on for a data sequence of a predetermined number of frames, and the operation of the fourth embodiment is performed by N systems. In addition, the reliability determination unit selects N ′ weight sequences (here, N ′: natural number, 1 ≦ N ′ <N) having a high reliability in the number of frames. After the reliability determination, for the remaining data series until the reliability determination is performed again at the above time interval, the second changeover switch 11 turns ON only the selected weight series switch, and switches the other weight series switches. Is OFF, and the same operation as that of the fourth embodiment is performed in the N ′ system using the selected N ′ weight sequences.

(Modification 2 of the eighth embodiment)
In the above description, the likelihood averaging unit 12 calculates the likelihood by a predetermined number of frames, Y1 frame (here, Y1: natural number, Y1 ≧ 1). , A method of selecting the minimum value, and a method of selecting the maximum value.

(Ninth embodiment)
(Configuration of Ninth Embodiment)
FIGS. 36A and 36B are block diagrams illustrating a reliability determination unit according to the ninth embodiment. The function blocks other than the reliability determination unit conform to the fourth embodiment and are omitted. The same parts as those in the fourth embodiment shown in FIGS. 31A and 31B are denoted by the same reference numerals.

(Operation of Ninth Embodiment)
Next, the operation of the reliability determination unit of the ninth embodiment will be described. The operation of the other functional blocks conforms to the fourth embodiment, and is omitted.

The RAKE-combined signal is input to the power calculator 13. The power calculator 13 calculates the power of the N systems after the RAKE combination, averages the power for a predetermined period, and inputs the calculated value to the reliability comparator 8.

On the other hand, the FEC decoding unit 5 decodes the error correction code for the RAKE-combined data sequence from the RAKE combining unit 4 and outputs decoded data of weight coefficients # 1 to #N. The CRC decoder 6 decodes the CRC using the CRC bits extracted from the data sequence decoded by the FEC decoder 5 to determine the presence or absence of a frame error, and outputs the determination result to the frame error number calculator 7. To enter.

The frame error number calculation unit 7 counts the number of frame errors existing in a predetermined Y frame (here, Y: natural number, Y ≧ 1), and inputs the counted number to the reliability comparison unit 8. .

The reliability comparison unit 8 selects, as an information output, a data sequence having the highest reliability than the power calculation value from among the systems having the least number of frame errors from the N frame error information output from the frame error number calculation unit 7. select.

(Effects of the Ninth Embodiment)
As described above, according to the ninth embodiment, the reception power after the RAKE combining is used for reliability determination together with the number of frame errors counted from the CRC decoding result, so that mutual determination factors are combined. Thus, strict reliability determination can be performed.

(Modification of the ninth embodiment)
In the above description, the processes from channel estimation to CRC decoding using the N weight sequences are always performed. However, the following modifications can reduce the load on the system.

変 形 A modification of the ninth embodiment can be configured by replacing the reliability determination unit of the modification of the fourth embodiment shown in FIGS. 31A and 31B with the ninth embodiment shown in FIGS. 36A and 36B.

{Circle around (2)} At every fixed period, all the second changeover switches 11 are turned on for a data sequence of a predetermined number of frames, and the operation of the fourth embodiment is performed by N systems. In addition, the reliability determination unit selects N ′ (N ′ <N) weight sequences having high reliability with the number of frames. After the reliability determination, for the remaining data series until the reliability determination is performed again at the above time interval, the second changeover switch 11 turns ON only the selected weight series switch, and switches the other weight series switches. Is OFF, and the same operation as that of the fourth embodiment is performed in the N ′ system using the selected N ′ weight sequences.

(Tenth embodiment)
(Configuration of Tenth Embodiment)
FIG. 37A and FIG. 37B are block diagrams illustrating a reliability determination unit according to the tenth embodiment. The function blocks other than the reliability determination unit conform to the fourth embodiment and are omitted. The same parts as those in the fourth embodiment shown in FIGS. 31A and 31B are denoted by the same reference numerals.

(Operation of Tenth Embodiment)
Next, the operation of the reliability determination unit of the tenth embodiment will be described. The operation of the other functional blocks conforms to the fourth embodiment, and is omitted.

The RAKE-combined signal is input to the SN ratio calculator 14. The S / N ratio calculator 14 calculates the S / N ratio of the signals after the RAKE combining of the N systems, averages them for a predetermined period, and inputs the calculated value to the reliability comparator 8.

On the other hand, the FEC decoding unit 5 decodes the error correction code for the RAKE-combined data sequence from the RAKE combining unit 4 and outputs decoded data of weight coefficients # 1 to #N. The CRC decoder 6 decodes the CRC using the CRC bits extracted from the data sequence decoded by the FEC decoder 5 to determine the presence or absence of a frame error, and outputs the determination result to the frame error number calculator 7. To enter.

The frame error number calculation unit 7 counts the number of frame errors existing in a predetermined Y frame (here, Y: natural number, Y ≧ 1), and inputs the counted number to the reliability comparison unit 8. .

The reliability comparing unit 8 outputs a data sequence having the highest reliability from the SN ratio calculation value from among the systems having the least number of frame errors from the N systems of frame error information output from the frame error number calculating unit 7. Select as

(Effects of the Tenth Embodiment)
As described above, according to the tenth embodiment, the SN ratio after the RAKE combining is used for reliability determination together with the number of frame errors counted from the CRC decoding result, so that mutual determination factors are combined. Thus, strict reliability determination can be performed.

(Modification of Tenth Embodiment)
In the above description, the processes from channel estimation to CRC decoding using the N weight sequences are always performed. However, the following modifications can reduce the load on the system.

変 形 A modification of the tenth embodiment can be configured by replacing the reliability determination unit of the modification of the fourth embodiment shown in FIGS. 31A and 31B with the tenth embodiment shown in FIGS. 37A and 37B.

{Circle around (2)} At every fixed period, all the second changeover switches 11 are turned on for a data sequence of a predetermined number of frames, and the operation of the fourth embodiment is performed by N systems. In addition, the reliability determination unit selects N ′ weight sequences (here, N ′: natural number, 1 ≦ N ′ <N) having a high reliability in the number of frames. After the reliability determination, for the remaining data series until the reliability determination is performed again at the above time interval, the second changeover switch 11 turns ON only the selected weight series switch, and switches the other weight series switches. Is OFF, and the same operation as that of the fourth embodiment is performed in the N ′ system using the selected N ′ weight sequences.

(Summary of Fourth to Tenth Embodiments)
FIG. 38 is a diagram illustrating a general concept in the fourth to tenth embodiments described so far. In this figure, 30 is a despreading unit, 40 (including 40-1 to 40-N) is a receiving unit, 50 (including 50-1 to 50-N) is a quality measuring unit, and 60 is a quality comparing / determining unit. , 70 are output changeover switches.

That is, in a demodulation device compliant with the CDMA system by direct spreading, a pilot signal is weighted and averaged using a plurality of predetermined weight sequences to obtain a channel estimation value. Then, the received data is demodulated using the obtained channel estimation value (40), and the reliability of the plurality of demodulated data is determined to select one of the highest quality output data (50, 60). , 70).

Also, for a certain period, it is possible to select some weight sequences based on the reliability determination result of the demodulated data sequence. In this case, thereafter, demodulation is performed only with the selected weight sequence.

Note that the channel estimation units 120, 220, and 320 in the first to third embodiments can be used as the channel estimation unit 2 in the fourth to tenth embodiments.

FIG. 3 is a diagram illustrating an example of a frame configuration of a signal received by the demodulation device according to the first embodiment of the present invention. FIG. 4 is a diagram for explaining a channel estimation method by the demodulation device according to the first embodiment of the present invention. FIG. 2 is a block diagram illustrating a configuration example of a demodulation device according to the first embodiment of the present invention. FIG. 2 is a block diagram illustrating a configuration example of a channel estimation unit according to the first embodiment of the present invention. FIG. 3 is a block diagram illustrating a configuration example of a fading frequency determination unit according to the first embodiment of the present invention. It is a figure showing the example of calculation of a channel estimation value. FIG. 11 is a diagram illustrating an example in which a data symbol in one slot is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. FIG. 11 is a diagram illustrating an example in which a data symbol in one slot is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. FIG. 11 is a diagram illustrating an example in which a data symbol in one slot is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. FIG. 11 is a diagram illustrating an example in which a data symbol in one slot is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. FIG. 3 is a diagram for explaining the concept of fading frequency determination. FIG. 3 is a diagram for explaining the concept of fading frequency determination. FIG. 9 is a diagram illustrating a result obtained by a computer simulation of a measurement value with respect to a measurement time using a fading frequency as a parameter. FIG. 13B is a diagram showing the relationship between FIG. 13A and FIG. 13B. It is a block diagram showing another example of composition of the fading frequency judgment part concerning a 1st embodiment of the present invention. It is a block diagram showing another example of composition of the fading frequency judgment part concerning a 1st embodiment of the present invention. FIG. 4 is a diagram for describing an example of determining a fading frequency. FIG. 4 is a diagram illustrating an example in which a transmission rate of a data channel is different from a transmission rate of a control channel. It is a figure showing an example of the frame composition of the signal which the demodulation device concerning a 2nd embodiment of the present invention receives. It is a block diagram showing an example of composition of a demodulation device concerning a 2nd embodiment of the present invention. FIG. 11 is a diagram illustrating an example in which a data symbol in one slot is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. FIG. 11 is a diagram illustrating an example in which a data symbol in one slot is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. FIG. 11 is a diagram illustrating an example in which a data symbol in one slot is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. FIG. 11 is a diagram illustrating an example in which a data symbol in one slot is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. It is a figure showing an example of the frame composition of the signal which the demodulation device concerning a 3rd embodiment of the present invention receives. It is a block diagram showing the example of composition of the demodulation device concerning a 3rd embodiment of the present invention. FIG. 14 is a block diagram illustrating a configuration example of a channel estimation unit according to a third embodiment of the present invention. FIG. 9 is a diagram illustrating an example in which a data symbol of a data channel is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. FIG. 9 is a diagram illustrating an example in which a data symbol of a data channel is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. FIG. 9 is a diagram illustrating an example in which a data symbol of a data channel is divided into a plurality of data symbol sections, and a channel estimation value is calculated for each data symbol section. FIG. 3 is a diagram for explaining the concept of fading frequency determination. FIG. 3 is a diagram for explaining the concept of fading frequency determination. FIG. 3 is an explanatory diagram of an example of a channel estimation method using a pilot signal. FIG. 30B is a diagram showing the relationship between FIG. 30A and FIG. 30B. It is a block diagram of a structure of the reliability determination part in 4th Embodiment. It is a block diagram of a structure of the reliability determination part in 4th Embodiment. FIG. 31B is a diagram showing the relationship between FIG. 31A and FIG. 31B. It is a block diagram of a structure of the reliability determination part in the modification of 4th Embodiment. It is a block diagram of a structure of the reliability determination part in the modification of 4th Embodiment. It is a block diagram of a structure of the reliability determination part in 5th Embodiment. It is a block diagram of a structure of the reliability determination part in 6th Embodiment. It is a block diagram of a structure of the reliability determination part in 7th Embodiment. FIG. 35B is a diagram showing the relationship between FIG. 35A and FIG. 35B. It is a block diagram of a structure of the reliability determination part in 8th Embodiment. It is a block diagram of a structure of the reliability determination part in 8th Embodiment. FIG. 36B is a diagram showing the relationship between FIG. 36A and FIG. 36B. It is a block diagram of a structure of the reliability determination part in 9th Embodiment. It is a block diagram of a structure of the reliability determination part in 9th Embodiment. FIG. 37B is a diagram showing the relationship between FIG. 37A and FIG. 37B. It is a block diagram of a structure of the reliability determination part in 10th Embodiment. It is a block diagram of a structure of the reliability determination part in 10th Embodiment. It is the figure which drew the high-order concept in 4th-10th embodiment.

Explanation of reference numerals

1, 30 despreading unit 2-1 to 2-N channel estimating unit 3-1 to 3-N multiplying unit 4-1 to 4-N RAKE combining unit 5, 5-1 to 5-N FEC decoding unit 6-1 6-N CRC decoding unit 7-1 to 7-N Frame error calculation unit 8A to 8H Reliability comparison unit 9 Reliability determination unit 10 First changeover switch 11 Second changeover switch 12-1 to 12-N Yu Degree averaging unit 13-1 to 13-N power calculation unit 14-1 to 14-N SN ratio calculation unit 40-1, 40-2 reception unit 50-1, 50-2 quality measurement unit 60 quality comparison / judgment unit 70 output changeover switch 102, 302 matched filter for data channel 104, 126, 128, 130, 163-1, 163-2, 204, 304, 326, 328, 330 delay section 106 matched filter for control channel 108, 132 134, 136, 208, 308, 332, 334, 336 Multiplication unit 110, 210, 310 RAKE combining unit 120, 220, 320 Channel estimation unit 122 Slot synchronization detection unit 124, 324 Pilot symbol averaging unit 138, 338 Weighting coefficient control Units 140, 340 Addition unit 150, 350 Fading frequency determination unit 152, 162 Normalization unit 154, 164-1, 164-2 Inner product value calculation unit 156, 166-1, 166-2 First averaging unit 158, 168- 1, 168-2 Second averaging unit 160 Threshold judgment unit 169 Difference calculation unit 170 Judgment unit 202 Matched filter 306 Matched filter for pilot channel

Claims (16)

A fading frequency determination device,
Inner product value calculating means for calculating an inner product value of a pilot symbol time-multiplexed on the control channel parallel-multiplexed with the data channel;
Determining means for determining a fading frequency based on the inner product value calculated by the inner product value calculating means. The fading frequency determination device according to claim 1,
The inner product value calculating means,
Normalizing means for normalizing an average value of pilot symbols in each of two slots of the control channel;
Inner product value calculation executing means for calculating an inner product value of an average value of two pilot symbols normalized by the normalizing means;
Inner product value averaging means for averaging the inner product value calculated by the inner product value calculation executing means over a plurality of slots of the control channel,
The determining means includes:
A fading frequency determination device, comprising: a determination executing unit that determines a fading frequency by comparing an inner product value averaged by the inner product value averaging unit with a threshold value. 3. The fading frequency determination device according to claim 2, wherein when the inner product value averaged by the inner product value averaging means is larger than a certain value, two slots of the control channel farther apart from each other. For the average value of the pilot symbols in each of the above, perform the normalization, the inner product value calculation and the inner product value averaging, and compare the obtained averaged inner product value with a threshold corresponding to the farther interval. A fading frequency determination device for determining a fading frequency. The fading frequency determination device according to claim 1,
The inner product value calculating means,
Normalizing means for normalizing an average value of pilot symbols in each of two slots of the control channel for each of multipaths used for RAKE combining;
Inner product value calculation executing means for calculating an inner product value of an average value of two pilot symbols normalized by the normalizing means for each of the multipaths;
First inner product value averaging means for averaging the inner product value of each of the multipaths calculated by the inner product value calculation executing means,
Second inner product value averaging means for averaging the inner product value averaged by the first inner product value averaging means over a plurality of slots of the control channel,
The determining means includes:
A fading frequency determination device, comprising: a determination executing unit that determines a fading frequency by comparing an inner product value averaged by the second inner product value averaging unit with a threshold. The fading frequency determination device according to claim 1,
The inner product value calculating means,
Normalizing means for normalizing an average value of pilot symbols in each of two slots of the control channel;
Inner product value calculation executing means for calculating two or more inner product values of the average value of the two pilot symbols normalized by the normalizing means while changing the inner product measurement interval;
For each inner product measurement interval, inner product value averaging means for averaging the inner product value calculated by the inner product value calculation executing means over a plurality of slots of the control channel,
The determining means includes:
A fading frequency determination device, comprising: a determination execution unit that determines a fading frequency using an inner product value for each inner product measurement interval averaged by the inner product value averaging unit. The fading frequency determination device according to claim 1,
The inner product value calculating means,
Normalizing means for normalizing an average value of pilot symbols in each of two slots of the control channel for each of multipaths used for RAKE combining;
Inner product value calculation executing means for calculating an inner product value of an average value of two pilot symbols normalized by the normalizing means for each of the multipaths by changing an inner product measurement interval, two or more;
First inner product value averaging means for averaging the inner product value of each of the multipaths calculated by the inner product value calculation execution means for each inner product measurement interval;
Second inner product value averaging means for averaging the inner product value averaged by the first inner product value averaging means over a plurality of slots of the control channel, for each inner product measurement interval,
The determining means includes:
A fading frequency determination device comprising: a determination execution unit that determines a fading frequency using an inner product value for each inner product measurement interval averaged by the second inner product value averaging unit. A fading frequency determination device,
Inner product value calculating means for calculating an inner product value of a pilot symbol in a channel in which the data symbol and the pilot symbol are time-multiplexed,
Determining means for determining a fading frequency based on the inner product value calculated by the inner product value calculating means. The fading frequency determination device according to claim 7,
The inner product value calculating means,
Normalizing means for normalizing an average value of pilot symbols in each of two slots of the channel;
Inner product value calculation executing means for calculating an inner product value of an average value of two pilot symbols normalized by the normalizing means;
Inner product value averaging means for averaging the inner product value calculated by the inner product value calculation executing means over a plurality of slots of the channel,
The determining means includes:
A fading frequency determination device, comprising: a determination executing unit that determines a fading frequency by comparing an inner product value averaged by the inner product value averaging unit with a threshold value. The fading frequency determination device according to claim 7,
The inner product value calculating means,
Normalizing means for normalizing an average value of pilot symbols in each of two slots of the control channel for each of multipaths used for RAKE combining;
Inner product value calculation executing means for calculating an inner product value of an average value of two pilot symbols normalized by the normalizing means for each of the multipaths;
First inner product value averaging means for averaging the inner product value of each of the multipaths calculated by the inner product value calculation executing means,
Second inner product value averaging means for averaging the inner product value averaged by the first inner product value averaging means over a plurality of slots of the channel,
The determining means includes:
A fading frequency determination device, comprising: a determination executing unit that determines a fading frequency by comparing an inner product value averaged by the second inner product value averaging unit with a threshold. The fading frequency determination device according to claim 7,
The inner product value calculating means,
Normalizing means for normalizing an average value of pilot symbols in each of two slots of the channel;
Inner product value calculation executing means for calculating two or more inner product values of the average value of the two pilot symbols normalized by the normalizing means while changing the inner product measurement interval;
For each inner product measurement interval, inner product value averaging means for averaging the inner product value calculated by the inner product value calculation executing means over a plurality of slots of the control channel,
The determining means includes:
A fading frequency determination device, comprising: a determination execution unit that determines a fading frequency using an inner product value for each inner product measurement interval averaged by the inner product value averaging unit. The fading frequency determination device according to claim 7,
The inner product value calculating means,
Normalizing means for normalizing the average value of pilot symbols in each of the two slots of the channel for each of the multipaths used for RAKE combining;
Inner product value calculation executing means for calculating an inner product value of an average value of two pilot symbols normalized by the normalizing means for each of the multipaths by changing an inner product measurement interval, two or more;
First inner product value averaging means for averaging the inner product value of each of the multipaths calculated by the inner product value calculation execution means for each inner product measurement interval;
Second inner product value averaging means for averaging the inner product value averaged by the first inner product value averaging means over a plurality of slots of the control channel, for each inner product measurement interval,
The determining means includes:
A fading frequency determination device comprising: a determination execution unit that determines a fading frequency using an inner product value for each inner product measurement interval averaged by the second inner product value averaging unit. A fading frequency determination device,
Inner product value calculating means for calculating an inner product value of pilot symbols of pilot channels multiplexed in parallel to the data channel,
Determining means for determining a fading frequency based on the inner product value calculated by the inner product value calculating means. The fading frequency determination device according to claim 12,
The inner product value calculating means,
Normalizing means for normalizing an average value of pilot symbols in each of two sections of the pilot channel;
Inner product value calculation executing means for calculating an inner product value of an average value of two pilot symbols normalized by the normalizing means;
Inner product value averaging means for averaging the inner product value calculated by the inner product value calculation executing means over a plurality of sections of the channel,
The determining means includes:
A fading frequency determination device, comprising: a determination executing unit that determines a fading frequency by comparing an inner product value averaged by the inner product value averaging unit with a threshold value. The fading frequency determination device according to claim 12,
The inner product value calculating means,
Normalizing means for normalizing an average value of pilot symbols in each of the two sections of the pilot channel for each of the multipaths used for RAKE combining;
Inner product value calculation executing means for calculating an inner product value of an average value of two pilot symbols normalized by the normalizing means for each of the multipaths;
First inner product value averaging means for averaging the inner product value of each of the multipaths calculated by the inner product value calculation executing means,
Second inner product value averaging means for averaging the inner product value averaged by the first inner product value averaging means over a plurality of sections of the pilot channel,
The determining means includes:
A fading frequency determination device, comprising: a determination executing unit that determines a fading frequency by comparing an inner product value averaged by the second inner product value averaging unit with a threshold. The fading frequency determination device according to claim 12,
The inner product value calculating means,
Normalizing means for normalizing an average value of pilot symbols in each of two sections of the pilot channel;
Inner product value calculation executing means for calculating two or more inner product values of the average value of the two pilot symbols normalized by the normalizing means while changing the inner product measurement interval;
For each inner product measurement interval, inner product value averaging means for averaging the inner product value calculated by the inner product value calculation executing means over a plurality of sections of the control channel,
The determining means includes:
A fading frequency determination device, comprising: a determination execution unit that determines a fading frequency using an inner product value for each inner product measurement interval averaged by the inner product value averaging unit. The fading frequency determination device according to claim 12,
The inner product value calculating means,
Normalizing means for normalizing an average value of pilot symbols in each of the two sections of the pilot channel for each of the multipaths used for RAKE combining;
Inner product value calculation executing means for calculating, for each of the multipaths, an inner product value of an average value of two pilot symbols normalized by the normalizing means, by changing an inner product measurement interval, two or more;
First inner product value averaging means for averaging the inner product value of each of the multipaths calculated by the inner product value calculation execution means for each inner product measurement interval;
Second inner product value averaging means for averaging the inner product value averaged by the first inner product value averaging means over a plurality of sections of the control channel, for each inner product measurement interval,
The determining means includes:
A fading frequency determination device comprising: a determination execution unit that determines a fading frequency using an inner product value for each inner product measurement interval averaged by the second inner product value averaging unit.

Images